Creation and Characterization of Synthetic Biological Tools for Modular Genetic Circuit Design

W&M ScholarWorks

Undergraduate Honors Theses Theses, Dissertations, & Master Projects

5-2019

Ethan Jones

Follow this and additional works at: https://scholarworks.wm.edu/honorstheses

Part of the Biology Commons, Biotechnology Commons, and the Systems Biology Commons

Recommended Citation Jones, Ethan, "Creation and Characterization of Synthetic Biological Tools for Modular Genetic Circuit Design" (2019). Undergraduate Honors Theses. Paper 1395. https://scholarworks.wm.edu/honorstheses/1395

This Honors Thesis is brought to you for free and open access by the Theses, Dissertations, & Master Projects at W&M ScholarWorks. It has been accepted for inclusion in Undergraduate Honors Theses by an authorized administrator of W&M ScholarWorks. For more information, please contact [email protected].

Abstract

The core goal of synthetic biology is to obtain predictable and practical control over the various properties and behaviors of biological systems. To enable this control, new synthetic biological methods are expected to be rigorously characterized as well as usable in a broad variety of systems. Commonly, new synthetic biological methods focus on the creation of novel modular genetic parts and genetic parts-based schemes to control or tune various properties of genetic circuits. However, despite the advances that have been made in modular parts-based control over the last twenty years, there are currently few modular parts-based methods that allow the control of dynamic properties of genetic circuits. Here, I seek to ameliorate this issue by developing a modular genetic parts-based method for the control of the dynamic property of genetic circuit response time. Additionally, another current problem of genetic parts is that many genetic parts have had their modularity impaired by genetic context resulting from promoter junction interaction. While there are currently some methods that allow for the ‘insulation’, or prevention of this modularity impairing genetic context, there has been little to no investigation of what if any secondary effects might arise from the use of these methods. To address this, I investigate the secondary effects of the commonly used insulator RiboJ, showing that it potentially compromises modularity by increasing the expression of insulated genes.

Table of Contents Preface ...... 4 Introduction...... 6 Chapter I: Characterization of RiboJ ...... 10 Introduction ...... 10 Discussion ...... 16 Methods ...... 18 Author Contributions ...... 22 Chapter II: A Control System for Response Time ...... 23 Introduction ...... 23 Results ...... 24 Discussion ...... 29 Methods ...... 30 Discussion ...... 35 Appendix I- Supplementary Figures ...... 36 Appendix II- RiboJ construct sequences ...... 42 Bibliography ...... 48

Preface

The scientific work in this thesis represents the culmination of multiple iGEM team projects, two of which I served as the student team leader. Given that the work is necessarily collaborative, below I delineate the contributions of other students as well as myself to the work described in this thesis. The work contained within Chapter I details the results of a characterization of the genetic insulator RiboJ which was published in the Journal of Biological Engineering with myself and Kalen Clifton as co-first authors

(Clifton et al., 2018). This idea for this characterization project was first conceived during the 2016 iGEM project by Andrew Halleran and John Marken. After the completion of iGEM 2016, Kalen Clifton and I proceeded to work on this project, with some advising from John and Andrew. While Kalen and I designed, coordinated and ran the major experiments, we were also assisted at various stages by the other co-authors.

A full summary of author contributions can be found in the author contributions section of Chapter I.

Chapter II details the creation of modular genetic parts-based system for response time control. This chapter is based upon the work performed as part of iGEM

2017 of which I was team leader. The project was originally conceived by myself and

John Marken, and the overall planning, implementation, design and organization of the project was performed by myself. While the conceptual, organizational and experimental details were largely determined by myself with the advising of John Marken, the cloning of constructs and experimental work was split evenly amongst the iGEM 2017 team.

After the conclusion of iGEM, Callan Monette, John Marken and myself have continued

4 to work on the project and a preprint of our work is available on Biorxiv (Jones et al.,

2018). A full summary of author contributions can be found in the author contributions section of Chapter II.

Introduction

Synthetic biology is an interdisciplinary field that is focused on applying engineering principles to the design, construction and modification of biological systems and tools. Although the field of synthetic biology is less than 20 years old, advances made in the field have led to the creation of numerous solutions to real world problems.

Advancements from synthetic biology have, among other things, increased researchers’ ability to design biomaterials, enhanced the use of biological systems for industrial production, enabled deeper foundational research into cellular processes, and yielded new treatments and tests in medicine and diagnostics.

Some of the most straightforward applications of synthetic biology are found in bio-production, where synthetic biology can be used to enhance or substitute for existing bio-production processes, as well as to create novel biological products. For example, existing bioprocesses have been enhanced by the engineering of recombinant enzymes for traits such as improved catalytic efficiency (Cahn, Baumschlager,

Brinkmann-Chen & Arnold, 2015), greater enantioselectivity (Peters, Meinhold, Glieder

& Arnold, 2003) and compatibility with organic solvents (Wong, Arnold & Schwaneberg,

(2004). Bio-production processes have also benefited from synthetic biological advances in genome editing (Ran et al., 2015; Jusiak, Cleto, Perez-Piñera & Lu, 2016;

Hu et al., 2018), which have enabled currently used organisms to be modified for higher production rates (Luo, Zeng., Du, Chen & Zhou, 2019), the elimination of unwanted side-products (Rogers, J., Guzman, C., Taylor, N., Raman, S., Anderson, K., & Church,

G. (2015), the creation of recombinant metabolic pathways (Antonovsky et al., 2016),

6 and the production of proteins containing unnatural amino acids (Lajoie et al.,2013; Ho et al., 2016; Wannier et al., 2018). Through the use of directed evolution and the engineering of enzymes, synthetic biology has enabled the use of novel reactions such as carbon-silicon bond formation, which are neither accessible through traditional synthetic chemistry nor found in any known natural enzyme (Kan, Lewis, Chen, &

Arnold, 2016).

Synthetic biology and synthetic biological tools have also been used in medical fields to create treatments and diagnostics. The most prominent example of this usage is the use of synthetic biological tools to develop chimeric antigen receptor T-cell (CAR

T-cell) based therapies for cancer treatment. These CAR T-cell therapies, the first of which was approved by the FDA in only 2017, show great potential for the treatment of currently intractable cancers, and it is for this reason that well over 200 CAR T-cell based therapies are currently in clinical trial (Miliotou & Papadopoulou, 2018). Beyond

CAR T-cell therapy, synthetic biology has also been used to develop lower cost and more accurate diagnostic assays (Harrington et al., 2018) as well as played a crucial role in the development of important “biologic” drugs. Given that biologic drugs such as adalimumab and abciximab now make up the majority of new pharmaceuticals (Lowe,

2018), it is likely that synthetic biological methods for the development, enhancement and characterization of potential new biologics will only continue to grow.

Although synthetic biology has already made a large impact on the medical field, it has an even greater potential in the long term. Drug delivery methods, synthetic organs and tissues, as well as gene therapies for diseases such as cystic fibrosis

(Marangi & Pistritto, 2018) and muscular dystrophy (Nelson et al., 2018; Min et al.,

2019) are all active areas of research. While some of these future medical applications require a great deal of foundational biological research before they come to fruition, others are likely to make an impact in the near future. Indeed, a clinical trial for Leber's congenital amaurosis, a genetic disease that causes blindness, began in December

2018 and represents the first FDA approved clinical trial for an in body CRISPR-based gene editing method (Sheridan, 2018).

In addition to these industrial and medical applications, synthetic biological tools have been used to answer fundamental biological questions. For instance, the development and improvement of optogenetic tools for gene expression

(Baumschlager, Aoki & Khammash, 2017; Zhao et al, 2018), ion channel activation

(Boyden, Zhang, Bamberg, Nagel & Deisseroth, 2005), phase separation (Shin et al.,

2017), and chromatin structure (Rege et al., 2018) has enabled the investigation of molecular mechanisms in a temporally and spatially precise manner. Similarly, genome editing tools such as CRISPR/Cas9 and TALENs have enabled the examination of biological questions in fields such as development (Xue, Tian, Fujii, Kladwang, Das &

Barna, 2015), genetics (Shalem et al., 2013; Wang et al., 2014) and immunology

(Hochheiser, Kueh, Gebhardt & Herold, 2018).

At the foundation of all of these different advances and future directions is the core concept of control. Fundamentally, the field of synthetic biology is centered around gaining and exploiting new methods for the control of biological systems, as greater ability to control biological systems enables greater ability to engineer novel functionality

8 of biological systems. The goal of this thesis is to further the aims of synthetic biology by: (1) improving our understanding of the properties of existing tools for genetic circuits: (2) designing modular and orthogonal methods of control for dynamical properties of genetic circuits, and (3) improving the practice of synthetic biology at

William & Mary by developing repeatable, in depth, and conceptually complete standardized protocols for synthetic biological methods.

Chapter I: Characterization of RiboJ

Introduction

One of the most common ways that synthetic biologists control biological systems is through the use of genetic circuits: genetic programs composed of standardized sequences of DNA. Ideally, each of these constituent sequences called

‘genetic parts’ should be standardized, as well as modular and well characterized. That is, a given genetic part should be able to be used with any other genetic part and behave as previously characterized regardless of the genetic parts it is used with. Over the last fifteen years, a great deal of effort has gone into the construction of standardized genetic parts, and currently large libraries of standardized and well characterized genetic parts exist (iGEM.org). However, despite the careful characterization of genetic parts from these standardized registries, many genetic circuits composed of these parts do not behave in a predictable manner and are instead influenced by the sequences of their neighboring genetic parts or “genetic context” (Lou,

Stanton, Chen, Munsky, & Voigt, 2012).

One major source of predictability-disrupting genetic context results from the usage of synthetic hybrid promoters: human designed promoters that combine the regulatory elements of multiple natural promoters to obtain novel functionality (Lou et al., 2012). While these synthetic hybrid promoters are quite useful, their functionality often depends on regulatory elements located after the transcriptional start site. When these regulatory sequences are transcribed, this leads to the creation of so called “RNA leaders”, additional unintended nucleotides on the 5’ end of the transcript. The presence

10 of these RNA leaders has been shown to alter gene expression due to altered stability and secondary structure due to interactions with downstream transcript sequences (Lou et al., 2012). Importantly, the precise nature of these RNA leader interactions is dependent on the specific sequence of the construct, meaning that a given promoter’s

RNA leaders will have wildly different effects for different genetic circuits (Figure 1.1a, b).

Figure 1.1: Schematic depicting role of RiboJ insulation on transcripts. Constructs in (a) and (b) have same coding region and are identical at the DNA level except for different promoters. However, the constructs in (a) and (b) result in different transcripts. Construct (b) has a synthetic promoter that contains an internal transcriptional start site leading to the inclusion of additional sequences in the transcript. This 5’ RNA leader can affect the stability of the RNA and result in different translational expression properties as well. The same constructs depicted in (a) and (b) are shown in (c) and d, except with the addition of the insulator RiboJ. Following transcription, the ribozyme RiboJ self-cleaves, resulting in standardized and identical transcripts

However, constructs can be designed to prevent the effect of these RNA leaders by including genetic insulators, which insulate genetic parts from neighboring sequences. One of the most commonly used genetic insulators is RiboJ, a self-cleaving ribozyme whose insulation ability was first described by Lou et al. (2012). In their paper,

Lou et al. sought to characterize the transfer function of NOT gates in which the output of one promoter repressed the activity of another. However, the investigators found that unanticipated promoter driven effects caused by variable RNA leaders lead to quantitatively different input-output responses between the characterized NOT gates.

Lou et al. then showed that the introduction of RiboJ immediately downstream of the promoter lead to equivalent transfer functions irrespective of the promoters used, thus insulating the circuit from promoter-dependent effects.

Based upon the work of Lou et al., RiboJ is now used as a standard tool to combat interference at the junction between promoters and 5’ untranslated regions

(UTR). A given genetic circuit can be insulated simply by placing the 75 nucleotide

RiboJ sequence at the junction of a promoter and 5’ UTR. Then, during post- transcriptional processing, RiboJ self-cleaves, removing upstream sequence, as well as any RNA leaders. After this processing, the hairpin containing sequence from the uncleaved region of RiboJ will be the only sequence upstream of the 5’ UTR, standardizing the 5’ end of any insulated transcript regardless of the promoter used.

This standardization of the 5’ end of each transcript prevents any secondary effects resulting from RNA leaders, leading to more consistent behavior of promoters between different constructs (Figure 1.1c, d.)

Despite RiboJ’s widespread use, there has been little to no characterization of what, if any, effects the presence of RiboJ itself has upon gene expression. Since RiboJ is used in the construction of genetic circuits to increase predictability and accuracy, any effects on gene expression resulting from its use could prove detrimental. In this chapter

I detail the characterization of the impact of insulation with RiboJ on the gene expression of a set of constitutive promoter constructs, finding that RiboJ consistently increases the expression of insulated genes at both a protein and RNA level.

Results

To test the effects of RiboJ on promoter gene expression, we assembled a set of

24 pairs of gene expression constructs on the plasmid backbone psb3K3 using Gibson

Assembly (NEB). Each construct consisted of one of 24 commonly used synthetic constitutive promoters (Appendix I), coupled to a B0034 superfolder-GFP (sfGFP) (Lou et al., 2012) expression cassette either with or without RiboJ. These constructs were then transformed into the protein expression strain BL21, and measurements of gene expression at the protein and RNA levels were determined using flow cytometry and reverse transcription digital droplet qPCR (ddPCR).

For each pair of promoter constructs, we found that the absolute fluorescence values of constructs insulated with RiboJ was consistently elevated with respect to the corresponding uninsulated construct (Figure 1.2a). This elevation spanned from approximately 2- to 10-fold (Figure 1.2b), and appeared to exhibit bimodality dependence on the strength of the insulated promoter. For 18 out of our assay’s 19 strongest promoters we found that RiboJ insulation lead to an approximately 8-fold average increase in gene expression, while by comparison the remaining weaker promoters increased gene expression by an average of around 4-fold.

Figure 1.2: a. Absolute fluorescence of constructs denoted by BioBrick ID, with (blue) and without (black) RiboJ insulation as measured by calibrated flow cytometry. Each dot represents the geometric mean fluorescence of n > 10,000 cells. b. Fold change in fluorescence of constructs when insulated with RiboJ. Bars represent the fold change in the mean fluorescence across replicates, and dots represent all pairwise fold changes between replicates. The dashed line and grey region indicate one geometric SD factor around the geometric mean of a null fold change distribution computed from the fluorescence data

Next, since many genetic circuits utilize RNA molecules instead of proteins as their output (Qi, Haurwitz, Shao, Doudna & Arkin, 2012), we characterized what if any effect RiboJ insulation had on gene expression at the RNA level. To do so, we used reverse transcription ddPCR to characterize the relative transcript abundance of sfGFP using the CysG as an endogenous reference (Peng, Stephan, Hummerjohann &

Tasara, 2014). We found that constructs insulated with RiboJ showed a two-fold average increase in sfGFP transcript abundance (Figure A1.1.2), whereas there was no fold change observed in the endogenous reference on average (Figure A1.1.3). We compared the distribution of fold change after insulation of sfGFP to the distributions of

CysG and a null model (Figure 1.3, Figure A1.4), and found that the mean fold change in transcripts was far greater, implying that the observed increase in transcript abundance is due to RiboJ.

Figure 1.3: Fold change in the transcript abundance of CysG, sfGFP, and null distribution when promoter constructs are insulated with RiboJ. P values were calculated using Welch’s one-tailed t-tests with hypotheses sfGFP > Null (p = 9.41e-10) and sfGFP > CysG (p = 4.94e-09). For the comparison of Null and CysG, Welch’s two-tailed t-test was used (p = 0.37). Dots represent all pairwise ratios of replicates (9 dots per promoter)

Discussion

Genetic insulators such as RiboJ are extremely valuable and versatile synthetic biological tools which can be used to improve standardization and consistency of genetic parts. Despite their importance, many of these vital tools lack characterization of

16 their effect on genetic circuits other than insulation. In this chapter, I detail our attempts to alleviate this lack of characterization by providing a quantitative characterization of one of the most commonly used genetic insulators, RiboJ. Our findings indicate that the use of RiboJ as an insulator consistently increases gene expression on both the protein and RNA level. This increase may lead to genetic circuits with inconsistent behavior if not accounted for.

Although our characterization revealed that RiboJ increases gene expression, the mechanism for this increase is still undetermined. Given that the development of genetic insulators is an active area of research (Mutalik et al., 2013; Davidsohn et al.,

2014), future work focusing on the mechanisms by which RiboJ exerts its effects on gene expression may prove valuable. While our work was not focused on the determination of the mechanism of RiboJ’s gene expression effects, our characterization does hint at potential mechanistic explanations. One potential explanation is that insulation with RiboJ increases transcript abundance, leading to a corresponding increase in protein expression. This increase in transcript abundance could be due to increased mRNA stability resulting from the short hairpin sequence left on the 5’ end of insulated transcripts after RiboJ’s self-cleavage, as terminal hairpins have been found to increase stability in other systems (Carrier & Keasling, 1997).

In support of this explanation, we found that sfGFP transcript abundance was well correlated with sfGFP protein expression, which could imply that increased sfGFP transcript abundance is driving increased sfGFP protein expression (Figure A1.5).

However, we found little correlation when we compared the fold change in transcript

17 abundance and protein expression between each insulated/uninsulated construct pair

(Figure A1.6). This poor correlation could potentially imply that translational processes instead of or as well as transcriptional abundance could be driving the increase in protein production. One way that RiboJ might alter the translation of transcripts could be that the post-cleavage hairpin of RiboJ increases exposure of the transcript’s RBS, increasing ribosome binding and hence protein production.

Given the importance of predictable and standardized genetic parts, future work is needed to determine the design constraints of ribozyme based genetic insulators.

While other methods such as CRISPR-based insulation are beginning to show promise, ribozyme-based insulators are currently the most widely utilized, as they are far simpler in implementation as well as much more modular. As ribozyme-based insulators are likely to be the standard of insulation in the near future, it is imperative that we research whether it is possible to engineer a ribozyme that possesses insulator effects without gene expression modification as well as methods to minimize or accommodate any irremovable gene expression effects.

Methods

Construction of library:

For each promoter we created two measurement constructs. The first contained the promoter, the ribosome binding site (RBS) Bba_B0034, and the double terminator

Bba_B0015. The second differed from the first only by the presence of RiboJ immediately upstream of the RBS sequence. Each of these 48 constructs (Appendix II) and a negative control (J23101 B0034 LacI B0015) were cloned from linear gene

18 fragments onto the low copy plasmid backbone pSB3K3 using NEB HiFi DNA assembly. Linear sequences for the promoters were obtained via PCR overlap of primers from IDT, and the remaining fragments were obtained from a template construct created from IDT gBlocks. Construct were transformed in 5-alpha Escherichia coli (E. coli) (NEB) and isolated via miniprep (NEB Monarch miniprep), and then confirmed by

Sanger sequencing.

Quantification of protein expression:

For assessment of RiboJ’s impact on expression level, each construct was transformed into chemically competent BL21 E. coli (NEB) using the manufacturer’s protocol. Then three distinct colonies were confirmed by colony PCR and grown overnight in 3 mL of LB containing 16 μg/mL kanamycin. After 12–14 h, saturated cultures were diluted 1:100 into 3 mL of M9 media containing 0.4% glucose and

16 μg/mL kanamycin and were grown to an Optical Density at 600 nm (OD600) of 0.5 as measured by plate reader (Biotek Synergy H1). For each sample, 1.5 mL of culture was pelleted at 6000x rpm, resuspended in 0.5 mL of Trizol, and stored at -20C immediately and then later moved to -80C for later RNA extraction. Of the remaining culture, 50 μl was filtered with 20 μm filters (CellTrics) into 0.5 mL of PBS and sfGFP expression was measured by flow cytometry for at least 10,000 cells per sample on the

FL1 channel of a Bio-Rad S3e cell sorter. Absolute fluorescence for each sample was calibrated using Spherotech Rainbow Calibration beads and the python package

FlowCal (Castillo-Hair et al., 2016).

RNA isolation and quantification:

Samples in Trizol were thawed on ice and homogenized using Lysing Matrix B

(MP Biomedical) along with a Bead Ruptor (Omni) for 1 min at speed 6, and the total

RNA of each sample was isolated using the MagMAX™ mirVana™ Total RNA Isolation

Kit (Applied Biosystems), with a 20 min DNAse step using the Turbo DNase enclosed in the kit. DNase activity was halted by the addition of 200 mM EDTA, and RNA was re- purified using the same MagMAX kit as before, quantified via a Nanodrop

Spectrophotometer, and stored in aliquots at -80C. Following the manufacturer’s protocol, 500 ng of total RNA was reverse transcribed into cDNA using an iScript cDNA

Synthesis Kit (Bio-Rad) and quantified via Nanodrop. Then Uroporphyrinogen-III C- methyltransferase (CysG) and sfGFP transcript levels were measured separately via

Taqman Assay (Thermofisher) using 1.0 or 0.1 ng of cDNA in a 20 μL volume reverse transcription digital droplet qPCR (ddPCR) (Bio-Rad) reaction. Positive droplet thresholds were set at 2800 for CysG and 4750 for sfGFP, and for each sample a no reverse transcriptase (RT) control was run with both assays; each plate also contained a no template control.

Analysis methods:

We used Flowcal to report fluorescence in Molecules of Equivalent Fluorophore

(MEF) instead of arbitrary units. Provided the fluorescence of Spherotech Rainbow

Calibration beads, Flowcal is able to determine the geometric mean of absolute fluorescence in MEF of at least 10,000 cells for each of our samples. Furthermore, we normalized the absolute fluorescence measured for our reporter constructs by

20 subtracting the absolute fluorescence of the negative control construct. All calculations were subsequently done with the normalized absolute fluorescence.

We created two (with RiboJ and without RiboJ) measurement constructs for each promoter and each of these two constructs had three biological replicates for which we measured fluorescence and transcript abundance. Therefore, each promoter has six protein and RNA measurements, three with RiboJ and three without RiboJ. We utilized two methods to represent the fold-change associated with RiboJ for each promoter. In the first method, the fold-change is a single value that is calculated as a ratio of the geometric mean of the three replicates with RiboJ and the geometric mean of the three replicates without RiboJ. This first method, referred to as “fold change of means,” is

% !"#"$"% equivalent to % , where for a given promoter xk is the measurement of the kth !&#&$&% replicate without RiboJ and yk is the measurement of the kth replicate with RiboJ.

Average fold-changes and ranges of fold-changes reported are determined using “fold change of means.” In the second method, the fold-change is represented by a set of nine values that correspond to all pairwise fold changes between replicates. These nine values are all possible ratios given by dividing a measurement of one of the replicates with RiboJ by a measurement of one of the replicates without RiboJ. This second

" method, referred to “pairwise fold-changes,” is summarized as ( ) : , ∈ {1, 2, 3}, 3 ∈ &*

{1, 2, 3}5 , where xk and yk are defined the same as above.

Additionally, we determined the distributions of fold-changes that are not associated with insulation with RiboJ for measurements of fluorescence and sfGFP

21 transcript abundance. This null distribution captures changes that would be introduced by natural expression variance and intrinsic noise in the measurement techniques. For

" & each promoter, these null pairwise fold-changes are computed by ( ) , ) : , ∈ {1, 2, 3}, 3 ∈ "* &*

{1, 2, 3}, , ≠ 35, that is dividing replicates from the same RiboJ condition and excluding identity ratios, where a replicate is divided by itself.

Author Contributions

Ethan Jones (EMJ) cloned constructs along with Kalen Clifton (KPC). EMJ designed experiments. EMJ developed trizol protocol and extracted samples with help from Sudip

Paudel (SP). EMJ performed flow cytometry measurements with help from KPC. Lidia

Epp (LE) performed ddPCR eperiments designed by EMJ and KPC. EMJ analyzed and normalized fluorescence and ddPCR. EMJ, KPC, Callen Monette (CEM) and John

Marken (JPM) created figures.

Chapter II: A Control System for Response Time

Introduction

As stated in Chapter I, genetic circuits are one of the main ways in which synthetic biologists control biological systems. Therefore, one important aim of synthetic biology is the development of methods that enable control of novel genetic circuit properties such as noise (Teo, Woo & Sarpeshkar, 2015; Chizzolini et al., 2017; Mundt,

Anders, Murray & Sourjik, 2018) or oscillation frequency (Stricker et al., 2008; Danino,

Mondragón-Palomino, Tsimring & Hasty, 2010). Currently, there exist many modular, genetic parts-based methods for control of various static properties of genetic circuits, such as steady state concentration (Mutalik et al., 2013) and circuit architecture (Taylor et al., 2015). However, there are considerably fewer methods that enable tuning of the dynamical properties of genetic circuits. Given that many important natural systems are inherently dynamic (Purvis & Lahav, 2013; Lin, Sohn, Dalal, Cai & Elowitz, 2015), this dearth of methods for the control of dynamical properties of genetic circuits represents a considerable hurdle for the creation of biological systems designed to emulate or integrate with relevant natural systems.

One fundamental dynamic property of genetic circuits is their response time, which is typically defined as the amount of time for the concentration of a circuit’s output to reach half of its steady-state value (Alon, 2006). Response time serves as valuable dynamical property in the design of effective genetic circuits, as it represents the rate at which a genetic circuit switches from an inactive state to an activated state. While there are currently some methods that enable control over the response time of genetic

23 circuits (Rosenfeld, Elowitz & Alon, 2002; Maeda & Sano, 2006; Gordley, Williams,

Bashor, Toettcher, Yan & Lim, 2016), these methods are not modular or genetic parts based, hindering their compatibility with existing design paradigms (Iverson, Haddock,

Beal & Densmore, 2016; Nielsen et al., 2016; Halleran, Swaminathan & Murray, 2018), and preventing their widespread use. In this chapter, I address this lack of compatibility by detailing the creation of a modular, orthogonal, genetic parts-based system to control the dynamical property of response time.

Results

To determine a general solution to the control of genetic circuit response time, we utilized a simple first order kinetic mathematical model of inducible gene expression described by Uri Alon (2006). In this model a gene produces an output protein (x) with a net production rate (α) and a degradation rate constant (γ) when expression is induced.

This forms the simple ordinary differential equation:

78 = ; − =8, 79

Which can be solved to obtain the concentration of the protein x as a function of time:

; 1 8(9) = @1 − D. = ABC

G With this equation, the steady state concentration of x can be determined to be 8 = . FF B

Circuit response time, t, can then be defined as the amount of time for the

24 concentration of protein x to reach half the steady state value, and the following equation can be obtained:

ln(2) H = . =

This expression reveals that an inducible gene’s response time is controlled only by the degradation rate of the gene product. Therefore, a modular parts-based approach that controls a gene’s degradation rate is sufficient to control its response time.

To create a modular parts-based approach to control response time via degradation rate, we utilized the Mesoplasma florum Lon protease (mf-Lon) protein degradation tag (pdt) system. The system is made up of two components. The first component is the mf-Lon protease, an E. coli orthogonal protease originally discovered by Gur and Sauer (2008), which is mechanistically similar to the E. coli ClpXP protease, possessing an affinity for a short C-terminal amino acid sequence. The second is a set of protein degradation tags which were developed by Cameron and Collins (2014).

Each tag possesses a distinct affinity for mf-Lon and therefore a unique degradation rate. Since this system can be used to tune the degradation rate of an arbitrary genetic circuit simply by adding a short 81 base pair (bp) sequence to the end of the desired gene, it represents an ideal method for modular parts-based control of degradation rate,

Before we began testing whether we could use this system to control response time, we first wanted to confirm that the system was capable of tuning degradation rate.

We created a series of anhydrotetracycline (ATc) inducible constructs which expressed the fast folding fluorescent reporter mScarlet-I (Bindels et al., 2016) as well as one of

25 five different pdts. Additionally, we created an Isopropyl β-D-1-thiogalactopyranoside

(IPTG) inducible mf-Lon expressing circuit on a separate lower copy plasmid backbone

(Figure 2.1a). We then co-transformed these plasmids, induced expression of mScarlet-

I and mf-Lon and measured their steady-state fluorescence values using flow cytometry.

From those values we were able to calculate the relative degradation rate of each pdt

(Figure 2.1b), and we found that these values were in line with the values reported by

Cameron and Collins (2014).

Figure 2.1: (a). Relative degradation rates measured in ATc inducible mScarlet-I pdt constructs. Each data point represents the population geometric mean of at least 10,000 cells of a distinct biological replicate; the line represents the geometric mean of replicates (b) Schematic of response time control system. An inducible fluorescent reporter x with a pdt is produced at rate α and is degraded by the E. coli orthogonal protease mf-Lon, increasing degradation rate and decreasing response time in a tag- specific fashion

Next, we attempted to determine if response time could be controlled by

degradation rate. We again induced expression of the reporter and mf-Lon and used

flow cytometry to measure fluorescence in 20-minute intervals until steady-state

fluorescence was reached. In accordance with the model, we found that increased

degradation rate of the reporter corresponded to a qualitatively faster steady state

(2.2a). Next, we further validated the model by comparing the relative degradation rates

of each pdt to experimental values for H, finding that the relationship appears to be

inverse as predicted by the model (2.2b).

Figure 2.2 (a) Measurements of steady-state normalized gene expression of ATc inducible mScarlet-I + pdt constructs for pdt #3 thru #3e (listed in order of increasing protease affinity). Data is shown for each construct until steady state is reached (at least two consecutive subsequent data points do not increase fluorescence). Each data point represents the geometric mean of 10,000 or more single cell measurements of three biological replicates. Shaded region represents one geometric standard deviation above and below the mean. (b) Comparison of calculated t1/2 vs relative degradation rates from 2.1.1b. t1/2 was defined as time at which each biological replicate's regression line reached half of steady state. The blue line represents an optical guide and is not fitted.

Having confirmed that our system was capable of tuning response time, we next

sought to extend its usability. Since degradation rate controls not only response time but

also steady-state concentration, one potential problem with our system might occur in

27 use cases that require tuning response time invariant to steady state concentration.

However, we speculated that this could be addressed by increasing the production rate of our output protein, as although response time is controlled by degradation rate alone, steady-state concentration is controlled by both degradation rate and production rate.

Therefore, our model predicts that our system is capable of increasing response time without altering steady-state concentration when combined with the tuning of production rate. We then tested this prediction by comparing the steady-state fluorescence as well as the H of one of our constructs at two different production rates. In accordance with our model, we found that by increasing the production rate of our reporter we were able to increase the steady-state fluorescence of our construct to a value similar to the no pdt case. Importantly, in accordance with our model, this increase in steady state concentration had no effect on response time (Figure 2.3). Since production rate can easily be tuned in any circuit by swapping ribosome binding sites or promoter variants

(Engler et al., 2014; Iverson et al., (2016); Halleran et al., 2018), our system should be easily able to control response time without changing steady state concentration.

Figure 2.3 Absolute (MEFL) (left) and steady-state normalized (right) values of mScarlet-I pdt constructs in multiple ATc induction conditions. As predicted by the model, increasing the production parameter (via increasing [ATc]) has minimal effect on the response time of the circuit (right), however increased [ATc] can restore the steady state value to the same value as seen in the no pdt (no degradation) condition.

Discussion

Modular genetic parts have been crucial in enabling the expansion of synthetic biology by allowing a simple and adaptable paradigm for the construction of genetic circuits for arbitrary functions. However, synthetic biology currently lacks modular genetic parts capable of tuning various dynamic properties of genetic circuits. Since many natural genetic circuits of interest are inherently dynamic, we sought to create a modular parts-based paradigm by which the important dynamic property of circuit response time could be controlled. To do so, we utilized and provided experimental confirmation of a fundamental mathematical model of gene expression that states that response time can be controlled by degradation rate. We implemented a modular genetic parts-based paradigm using the mf-Lon/pdt system, showing the response time can be controlled in a steady-state concentration invariant manner, and created modular

29 cloning constructs (BBa_K2333401-K2333406) that will enable easy use of our response time control system in a variety of settings.

While the system we developed represents a great leap forward in the ability to control response time using modular parts, there are a number of ways in which the system could be improved or extended. For instance, while the system has been characterized for use in E. coli, it should be possible to implement this or a similar system or in other bacterial species. Additionally, while our system is capable of decreasing the response time of genetic circuits, we should be able to utilize the same principle to increase the response time of a genetic circuit by hindering degradation.

Another question of particular of interest is whether or not a similar degradation-based paradigm for response time control could be created for use in Eukaryotes given their substantially more complex cellular structure and greater regulation of protein degradation. Given the importance of dynamics in areas such as neuroscience and development (Purvis & Lahav, 2013; Lin et al., 2015), better methods for the control of dynamical properties are of particular importance as they will greatly enhance our ability to understand and modify these systems.

Methods

Construct Design and Assembly

Amino acid sequences for pdts and mf-Lon were obtained from Cameron and

Collins 2014, manually codon optimized for E. coli using Integrated DNA Technology’s

(IDT) codon optimization tool, and then synthesized as gene blocks. The same process

30 was repeated for the fast folding red fluorescent protein mScarlet-I. Gene blocks were then cloned into the backbone pSB1C3. ATc inducible reporter constructs were cloned by using multipart Gibson Assembly to combine the pTet promoter (Bba_R0051), mScarlet-I and the various protein degradation tags (pdt) along with a constitutive TetR construct onto a pSB1C3 backbone. IPTG inducible mf-Lon constructs were cloned in a similar fashion, combining the pLac0-1 (Bba_R0011) promoter with mf-Lon along with a constitutive LacI construct onto the medium copy backbone pSB3K3. Construct sequences were deposited into the Biobrick registry (BBa_K2333427- K2333434)

Circuit response time characterization and readjustment characterization

ATc inducible reporter constructs were co-transformed with an IPTG inducible mf-Lon construct in the 10-Beta E. coli strain (New England Biolabs). Colonies were picked into M9 minimal media with 0.4% glycerol and grown overnight. In the morning, cultures were 1:100 diluted into fresh media and grown for 4-7 hours. Cell density was quantified, and cells were diluted to an optical density (OD) of 0.01 into fresh media containing 0.1mM IPTG and 50ng/mL ATc. Time points were taken from samples every

20 minutes on ice and then immediately thereafter, at least 10,000 (typically 20,000) single cell measurements were obtained using flow cytometry (FL3 channel Bio-Rad

S3e Cell Sorter). Additionally, 10,000 measurements of 8 peak Rainbow Calibration

Particles (Spherotech) were collected on each day in which measurement occurred.

Samples fluorescence was converted into MEFLs using the python package flowcal and the associated calibration beads. Data was analyzed and Figure 1b displays the inducible gene expression profile of a pulsatile IFFL with the mf-Lon protease serving as

31 an inhibitor and the inducers representing the input. For readjustment characterization, procedure was repeated while increasing the concentration of ATc to 100ng/mL.

Determination of steady state and calculation of t and relative degradation rate

Each timecourse was defined as reaching steady state at the first timepoint for which the two successive measurements were not above the fluorescence value at that timepoint. The steady-state value for each gene expression timecourse was defined as the fluorescence value obtained at the point the timecourse reached steady state. The measured timepoints were smoothed via a spline fit in Microsoft Excel to determine the response times, t, for each replicate by interpolation. Degradation rates for a given pdt were calculated as the ratio between the average steady-state fluorescence of the with- pdt conditions (across replicates) and the average steady-state fluorescence of the without-pdt conditions (across replicates) for that tag. In Figure 2b, the guide to the eye is a function of the form

K H = + N, L= where t and g are the response time and relative degradation rate, respectively, and A,

B, C are arbitrary constants.

Author Contributions

Ethan Jones (EMJ) designed constructs and construct assemblies. EMJ and John

Marken (JPM) conceived of the project. EMJ planned experiments and measured and analyzed flow cytometry data. EMJ. EMJ, Callen Monette (CEM), Cecilia Zhang (CZ),

Christine Li (CHL), Sejal Dhawan (SD), Alyssa Luz-Ricca (ALR) and Theresa Gibney

(TVG) cloned constructs, and prepped experiments. EMJ, TVG, JPM and CEM made figures.

Chapter III: Improving Sustainability in the BEL

Over the course of my research career, I have spent the majority of my time working in the Bioengineering Lab, and have accumulated a large amount of heretofore unwritten tips, tricks and general knowledge about both protocols and experimental design. Given the immense loss of knowledge that has accompanied the transitions of previous Synthetic Biology users out of the Bioengineering Lab, I have devoted the last few months to making, reworking and refining various laboratory protocols and standard operating procedures for the Bioengineering Lab. In the process, I have added as much information as is possible, including all of the “word of mouth tips” and their justification, as well as background, general purpose advice and the history of how the current procedure was decided upon and refined.

To make sure that my protocols would serve as both usable documents and stores of knowledge, I implemented standardized formatting, which enabled me to include the information needed to perform the protocol as well as any information that might one day be needed to develop the protocol further. Every protocol I have written uses standardized font, spacing, and sectioning, has strategic bolding of important numbers, and includes relevant references, observations, protocol history and any other non-directly protocol relevant information in footnotes. A selection of these protocols is attached, and it is my hope that these protocols will prevent the repetition of any previous mistakes.

Discussion

Although Synthetic biology is by its nature a fast-moving field, with new methods, parts and design principles constantly being developed, the protocols that I have developed should hopefully prove useful and adaptable for years to come. Ideally, future students in the Bioengineering Lab will be able to build off the foundational work in this thesis, as there are a number of potential future directions in which this work can be taken. Some of the more promising of these directions include the expansion of the response time control system to non-bacterial systems, the use of the underlying principles of the response time control system for more complex dynamical control systems, and the elucidation of the mechanism by which RiboJ effects gene expression.

Appendix I- Supplementary Figures

Figure A1.1 Fold Change in sfGFP Fluorescence associated with RiboJ Insulation: Dots indicate the pairwise fold change values computed between all replicates of a given construct. All constructs are pooled together into a single distribution. The null fold change distribution was computed from the sfGFP fluorescence data (see Chapter I methods). P-value was calculated from Welch’s one-tailed t-test with hypothesis sfGFP > Null (p=1.5e-153).

Figure A1.2: Counts and fold change for sfGFP transcripts: Top: sfGFP transcript counts for each biological replicate of each construct, obtained by ddPCR. Transcript count values for the negative control were <1. Bottom: RiboJ-associated fold change in sfGFP transcript count values. Black bars represent the fold change in the mean transcript count across replicates, and dots represent all pairwise fold changes between replicates. The grey region and dashed line indicate one geometric SD factor around the geometric mean of the null fold change distribution computed from the sfGFP transcript count data (Supplemental Methods).

Figure A1.3 Counts and fold change for CysG transcripts: Top: CysG transcript counts for each biological replicate of each construct, obtained ddPCR. Transcript count values that were less than 1 are not shown. Bottom: RiboJ-associated fold change in CysG transcript count values. Black bars represent the fold change in the mean transcript count across replicates, and dots represent all pairwise fold changes between replicates. Transcript count values <1 were excluded from fold change calculations.

Figure A1.4 RiboJ-associated fold change of mean transcript counts across replicates: Fold change in the transcript abundance of CysG and sfGFP when promoter constructs are insulated with RiboJ. Dots depict the fold change in the mean transcript count across the three replicates for a given construct (Supplemental Methods). All constructs are pooled into a single distribution. P-value was calculated from Welch’s one-tailed t-test with hypothesis sfGFP > CysG (p=0.006).

Figure A1.5 sfGFP fluorescence correlates with sfGFP transcript counts: Each dot depicts the relationship between a replicate fluorescence measurement and a replicate transcript count measurement, so each construct will appear 9 times on a given plot. Spearman’s rho = 0.61 (p = 1.8e-8) for the transcript count-fluorescence correlation in the RiboJ-insulated constructs (left), and Spearman’s rho = 0.67 (p = 4.3e-11) for the transcript count-fluorescence correlation in the non-insulated constructs (right).

Figure A1.6 sfGFP fluorescence fold change is generally higher than sfGFP transcript count fold change: Each dot depicts the relationship between the fold change in the geometric mean fluorescence and the fold change in mean transcript counts associated with RiboJ across all replicates for a given construct. For all but two promoters, the fold change in fluorescence is higher than the fold change in transcript count. As Spearman’s rho = 0.21 (p = 0.32), we cannot claim that there is a monotonic correlation between the variables.

Appendix II- RiboJ construct sequences

Table A2.1 (Promoter sequences) J23100 ttgacggctagctcagtcctaggtacagtgctagc

J23101 tttacagctagctcagtcctaggtattatgctagc

J23102 ttgacagctagctcagtcctaggtactgtgctagc

J23103 ctgatagctagctcagtcctagggattatgctagc

J23104 ttgacagctagctcagtcctaggtattgtgctagc

J23105 tttacggctagctcagtcctaggtactatgctagc

J23106 tttacggctagctcagtcctaggtatagtgctagc

J23107 tttacggctagctcagccctaggtattatgctagc

J23108 ctgacagctagctcagtcctaggtataatgctagc

J23109 tttacagctagctcagtcctagggactgtgctagc

J23110 tttacggctagctcagtcctaggtacaatgctagc

J23111 ttgacggctagctcagtcctaggtatagtgctagc

J23112 ctgatagctagctcagtcctagggattatgctagc

J23113 ctgatggctagctcagtcctagggattatgctagc

J23114 tttatggctagctcagtcctaggtacaatgctagc

J23115 tttatagctagctcagcccttggtacaatgctagc

J23116 ttgacagctagctcagtcctagggactatgctagc

J23117 ttgacagctagctcagtcctagggattgtgctagc

J23118 ttgacggctagctcagtcctaggtattgtgctagc

J23119 ttgacagctagctcagtcctaggtataatgctagc

J23150 tttacggctagctcagtcctaggtattatgctagc

J23151 ttgatggctagctcagtcctaggtacaatgctagc

R0010 caatacgcaaaccgcctctccccgcgcgttggccgattcattaatgcagctggcacgacaggtt tcccgactggaaagcgggcagtgagcgcaacgcaat

R0011 aattgtgagcggataacaattgacattgtgagcggataacaagatactgagcaca

Construct Design: Each construct of the two constructs below was assembled with each of the 24 promoter sequence (See supplementary sequences) at the site labeled xxx.

RiboJ Construct >Promoter Part XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

>RiboJ (from Lou et al. Supplement section V) Agctgtcaccggatgtgctttccggtctgatgagtccgtgaggacgaaacagcctctacaaataattttgtttaa

>BioBrick Scar (from Lou et al. Supplement section V) ACTAGA

> B0034 w/ Spacer (from Lou et al. Supplement section V) AAAGAGGAGAAATACTAG

>sfGFP (modified from Lou et al. Supplement section V) Atgcgtaaaggcgaagagctgttcactggtgtcgtccctattctggtggaactggatggtgatgtcaacggtcataagttttcc gtgcgtggcgagggtgaaggtgacgcaactaatggtaaactgacgctgaagttcatctgtactactggtaaactgccggta ccttggccgactctggtaacgacgctgacttatggtgttcagtgctttgctcgttatccggaccatatgaagcagcatgacttct tcaagtccgccatgccggaaggctatgtgcaggaacgcacgatttcctttaaggatgacggcacgtacaaaacgcgtgc ggaagtgaaatttgaaggcgataccctggtaaaccgcattgagctgaaaggcattgactttaaagaagacggcaatatcc tgggccataagctggaatacaattttaacagccacaatgtgtacattaccgcagataaacaaaaaaatggcattaaagcg aatttcaaaattcgccacaacgtggaggatggcagcgtgcagctggctgatcactaccagcaaaacactccaatcggtg atggtcctgttctgctgccagacaatcactatctgagcacgcaaagcgttctgtctaaagatccgaacgagaaacgcgatc atatggttctgctggagttcgtaaccgcagcgggcatcacgcatggtatggatgaactgtacaaatgatga

>BBa_B0015 Part-only sequence – double terminator ccaggcatcaaataaaacgaaaggctcagtcgaaagactgggcctttcgttttatctgttgtttgtcggtgaacgctctctact agagtcacactggctcaccttcgggtgggcctttctgcgtttata

>Complete Sequence XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXagctgtcaccggatgtgctttccggt ctgatgagtccgtgaggacgaaacagcctctacaaataattttgtttaaACTAGAAAAGAGGAGAAATACT AGatgcgtaaaggcgaagagctgttcactggtgtcgtccctattctggtggaactggatggtgatgtcaacggtcataagtt ttccgtgcgtggcgagggtgaaggtgacgcaactaatggtaaactgacgctgaagttcatctgtactactggtaaactgcc ggtaccttggccgactctggtaacgacgctgacttatggtgttcagtgctttgctcgttatccggaccatatgaagcagcatga cttcttcaagtccgccatgccggaaggctatgtgcaggaacgcacgatttcctttaaggatgacggcacgtacaaaacgc gtgcggaagtgaaatttgaaggcgataccctggtaaaccgcattgagctgaaaggcattgactttaaagaagacggcaa tatcctgggccataagctggaatacaattttaacagccacaatgtgtacattaccgcagataaacaaaaaaatggcattaa agcgaatttcaaaattcgccacaacgtggaggatggcagcgtgcagctggctgatcactaccagcaaaacactccaatc

44 ggtgatggtcctgttctgctgccagacaatcactatctgagcacgcaaagcgttctgtctaaagatccgaacgagaaacgc gatcatatggttctgctggagttcgtaaccgcagcgggcatcacgcatggtatggatgaactgtacaaatgatgaccaggc atcaaataaaacgaaaggctcagtcgaaagactgggcctttcgttttatctgttgtttgtcggtgaacgctctctactagagtc acactggctcaccttcgggtgggcctttctgcgtttata

No RiboJ Construct >Promoter Part XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

> B0034 w/ Spacer (from Lou et al. Supplement section V) AAAGAGGAGAAATACTAG

>sfGFP (modified from Lou et al. Supplement section V) Atgcgtaaaggcgaagagctgttcactggtgtcgtccctattctggtggaactggatggtgatgtcaacggtcat aagttttccgtgcgtggcgagggtgaaggtgacgcaactaatggtaaactgacgctgaagttcatctgtactact ggtaaactgccggtaccttggccgactctggtaacgacgctgacttatggtgttcagtgctttgctcgttatccgga ccatatgaagcagcatgacttcttcaagtccgccatgccggaaggctatgtgcaggaacgcacgatttcctttaa ggatgacggcacgtacaaaacgcgtgcggaagtgaaatttgaaggcgataccctggtaaaccgcattgagctg aaaggcattgactttaaagaagacggcaatatcctgggccataagctggaatacaattttaacagccacaatgtg tacattaccgcagataaacaaaaaaatggcattaaagcgaatttcaaaattcgccacaacgtggaggatggcag cgtgcagctggctgatcactaccagcaaaacactccaatcggtgatggtcctgttctgctgccagacaatcacta tctgagcacgcaaagcgttctgtctaaagatccgaacgagaaacgcgatcatatggttctgctggagttcgtaac cgcagcgggcatcacgcatggtatggatgaactgtacaaatgatga

>BBa_B0015 Part-only sequence – double terminator ccaggcatcaaataaaacgaaaggctcagtcgaaagactgggcctttcgttttatctgttgtttgtcggtgaacgct ctctactagagtcacactggctcaccttcgggtgggcctttctgcgtttata

>Complete Sequence XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXAAAGAGGAGAAATACTA Gatgcgtaaaggcgaagagctgttcactggtgtcgtccctattctggtggaactggatggtgatgtcaacggtca taagttttccgtgcgtggcgagggtgaaggtgacgcaactaatggtaaactgacgctgaagttcatctgtactact ggtaaactgccggtaccttggccgactctggtaacgacgctgacttatggtgttcagtgctttgctcgttatccgga ccatatgaagcagcatgacttcttcaagtccgccatgccggaaggctatgtgcaggaacgcacgatttcctttaa ggatgacggcacgtacaaaacgcgtgcggaagtgaaatttgaaggcgataccctggtaaaccgcattgagctg aaaggcattgactttaaagaagacggcaatatcctgggccataagctggaatacaattttaacagccacaatgtg tacattaccgcagataaacaaaaaaatggcattaaagcgaatttcaaaattcgccacaacgtggaggatggcag cgtgcagctggctgatcactaccagcaaaacactccaatcggtgatggtcctgttctgctgccagacaatcacta tctgagcacgcaaagcgttctgtctaaagatccgaacgagaaacgcgatcatatggttctgctggagttcgtaac cgcagcgggcatcacgcatggtatggatgaactgtacaaatgatgaccaggcatcaaataaaacgaaaggctc agtcgaaagactgggcctttcgttttatctgttgtttgtcggtgaacgctctctactagagtcacactggctcaccttc gggtgggcctttctgcgtttata

The negative control plasmid consists of J23101 B0034 (as above) LacI (Bba_C0012 without LVA tail) and B0015 (as above). tttacagctagctcagtcctaggtattatgctagcAAAGAGGAGAAATACTAGatggtgaatgtgaaacc agtaacgttatacgatgtcgcagagtatgccggtgtctcttatcagaccgtttcccgcgtggtgaaccaggccagccacgttt ctgcgaaaacgcgggaaaaagtggaagcggcgatggcggagctgaattacattcccaaccgcgtggcacaacaactg

46 gcgggcaaacagtcgttgctgattggcgttgccacctccagtctggccctgcacgcgccgtcgcaaattgtcgcggcgatt aaatctcgcgccgatcaactgggtgccagcgtggtggtgtcgatggtagaacgaagcggcgtcgaagcctgtaaagcg gcggtgcacaatcttctcgcgcaacgcgtcagtgggctgatcattaactatccgctggatgaccaggatgccattgctgtgg aagctgcctgcactaatgttccggcgttatttcttgatgtctctgaccagacacccatcaacagtattattttctcccatgaagac ggtacgcgactgggcgtggagcatctggtcgcattgggtcaccagcaaatcgcgctgttagcgggcccattaagttctgtct cggcgcgtctgcgtctggctggctggcataaatatctcactcgcaatcaaattcagccgatagcggaacgggaaggcga ctggagtgccatgtccggttttcaacaaaccatgcaaatgctgaatgagggcatcgttcccactgcgatgctggttgccaac gatcagatggcgctgggcgcaatgcgcgccattaccgagtccgggctgcgcgttggtgcggatatctcggtagtgggata cgacgataccgaagacagctcatgttatatcccgccgttaaccaccatcaaacaggattttcgcctgctggggcaaacca gcgtggaccgcttgctgcaactctctcagggccaggcggtgaagggcaatcagctgttgcccgtctcactggtgaaaaga aaaaccaccctggcgcccaatacgcaaaccgcctctccccgcgcgttggccgattcattaatgcagctggcacgacagg tttcccgactggaaagcgggcagtgaccaggcatcaaataaaacgaaaggctcagtcgaaagactgggcctttcg ttttatctgttgtttgtcggtgaacgctctctactagagtcacactggctcaccttcgggtgggcctttctgcgtttata

Bibliography

Alon, U. (2006) An Introduction to Systems Biology: Design Principles of Biological Circuits, 1st ed., pp. 18-22 Chapman and Hall/CRC.

Anderson promoter collection. http://parts.igem.org/Promoters/Catalog/Anderson. Accessed 20 April 2019.

Antonovsky, N., Gleizer, S., Noor, E., Zohar, Y., Herz, E., Barenholz, U., … Milo, R. (2016). Sugar Synthesis from CO2 in Escherichia coli. Cell, 166(1), 115–125. doi:10.1016/j.cell.2016.05.064

Baumschlager, A., Aoki, S., & Khammash, M. (2017). Dynamic Blue Light-Inducible T7 RNA Polymerases (Opto-T7RNAPs) for Precise Spatiotemporal Gene Expression Control. ACS Synthetic Biology. doi:10.1021/acssynbio.7b00169

Bindels, D., Haarbosch, L., Weeren, L., Postma, M., Wiese, K., Mastop, M., … Gadella Jr, T.W.J. (2016). mScarlet: a bright monomeric red fluorescent protein for cellular imaging. Nature Methods, 14(1), nmeth.4074. doi:10.1038/nmeth.4074

Boyden, E., Zhang, F., Bamberg, E., Nagel, G., & Deisseroth, K. (2005). Millisecond- timescale, genetically targeted optical control of neural activity. Nature Neuroscience, 8(9), nn1525. doi:10.1038/nn1525

Cahn, J. K. B., Baumschlager, Brinkmann-Chen, & Arnold, F. H. (2016). Mutations in adenine-binding pockets enhance catalytic properties of NAD(P)H-dependent enzymes. Protein Engineering, Design and Selection, 29(1), 31–38. doi:10.1093/protein/gzv057

Carrier, T. A. and Keasling, J. D. (1997), Engineering mRNA stability in E. coli by the addition of synthetic hairpins using a 5′ cassette system. Biotechnol. Bioeng., 55: 577- 580. doi:10.1002/(SICI)1097-0290(19970805)55:3<577::AID-BIT16>3.0.CO;2-D

Castillo-Hair, S. M., Sexton, J. T., Landry, B. P., Olson, E. J., Igoshin, O. A., & Tabor, J. J. (2016). FlowCal: A User-Friendly, Open Source Software Tool for Automatically Converting Flow Cytometry Data from Arbitrary to Calibrated Units. ACS synthetic biology, 5(7), 774–780. doi:10.1021/acssynbio.5b00284

Chizzolini, F., Forlin, M., Martín, N., Berloffa, G., Cecchi, D., & Mansy, S. (2017). Cell- free translation is more variable than transcription. ACS Synthetic Biology,6(4), 638– 647. doi:10.1021/acssynbio.6b00250

Clifton, K., Jones, E., Paudel, S., Marken, J., Monette, C., Halleran, A., … Saha, M. (2018). The genetic insulator RiboJ increases expression of insulated genes. Journal of Biological Engineering, 12(1), 23. doi:10.1186/s13036-018-0115-6

Danino, T., Mondragón-Palomino, O., Tsimring, L., & Hasty, J. (2010). A synchronized quorum of genetic clocks. Nature, 463(7279), 326. doi:10.1038/nature08753

Davidsohn, N., Beal, J., Kiani, S., Adler, A., Yaman, F., Li, Y., … Weiss, R. (n.d.). Accurate Predictions of Genetic Circuit Behavior from Part Characterization and Modular Composition. ACS Synthetic Biology, 4(6), 673–81. doi:10.1021/sb500263b

Engler, C., Youles, M., Gruetzner, R., Ehnert, T.-M., Werner, S., Jones, J., … Marillonnet, S. (2014). A Golden Gate Modular Cloning Toolbox for Plants. ACS Synthetic Biology, 3(11), 839–843. doi:10.1021/sb4001504

Gur, E., & Sauer, R. (2008). Evolution of the ssrA degradation tag in Mycoplasma: Specificity switch to a different protease. Proceedings of the National Academy of Sciences, 105(42), 16113–16118. doi:10.1073/pnas.0808802105

Gordley, R., Williams, R., Bashor, C., Toettcher, J., Yan, S., & Lim, W. (2016). Engineering dynamical control of cell fate switching using synthetic phospho- regulons.Proceedings of the National Academy of Sciences, 113(47), 13528–13533. doi:10.1073/pnas.1610973113

Halleran, A., Swaminathan, A., & Murray, R. (2018). Single day construction of multi- gene circuits with 3G assembly. ACS Synthetic Biology, 7(5), 1477–1480. doi:10.1021/acssynbio.8b00060

Harrington, L. B., Burstein, D., Chen, J. S., Paez-Espino, D., Ma, E., Witte, I. P., … Doudna, J. A. (2018). Programmed DNA destruction by miniature CRISPR-Cas14 enzymes. Science (New York, N.Y.), 362(6416), 839–842. doi:10.1126/science.aav4294

Ho, J., Reynolds, N., Rivera, K., Connolly, M., Guo, L.-T., Ling, J., … Söll, D. (2016). Efficient Reassignment of a Frequent Serine Codon in Wild-Type Escherichia coli.ACS Synthetic Biology, 5(2), 163–171. doi:10.1021/acssynbio.5b00197

Hochheiser, K. , Kueh, A. J., Gebhardt, T. & Herold, M. J. (2018), CRISPR/Cas9: A tool for immunological research. Eur. J. Immunol., 48: 576-583. doi:10.1002/eji.201747131

Hu, J., Miller, S., Geurts, M., Tang, W., Chen, L., Sun, N., … Liu, D. (2018). Evolved Cas9 variants with broad PAM compatibility and high DNA specificity. Nature,556(7699), 57. doi:10.1038/nature26155

Iverson, S., Haddock, T., Beal, J., & Densmore, D. (2016). CIDAR MoClo: Improved MoClo Assembly Standard and New E. coli Part Library Enable Rapid Combinatorial Design for Synthetic and Traditional Biology. ACS Synthetic Biology, 5(1), 99–103. doi:10.1021/acssynbio.5b00124

Jones, E., Monette, C., Marken, J., Dhawan, S., Gibney, T., Li, C., Liu, W., Luz-Ricca, A., Ren, X. Zheng, X., Saha, M. (2018). Modular, part-based control of gene expression response time using protein degradation tags. bioRxiv 482331; doi: https://doi.org/10.1101/482331

Jusiak, B., Cleto, S., Perez-Piñera, P., & Lu, T. (2016). Engineering Synthetic Gene Circuits in Living Cells with CRISPR Technology. Trends in Biotechnology, 34(7), 535– 547. doi:10.1016/j.tibtech.2015.12.014

Kan, S., Lewis, R., Chen, K., & Arnold, F. (2016). Directed evolution of cytochrome c for carbon–silicon bond formation: Bringing silicon to life. Science,354(6315), 1048–1051. doi:10.1126/science.aah6219

Lajoie, M., Rovner, A., Goodman, D., Aerni, H.-R., Haimovich, A., Kuznetsov, G., … Isaacs, F. (2013). Genomically Recoded Organisms Expand Biological Functions. Science, 342(6156), 357–360. doi:10.1126/science.1241459

Lin, Y., Sohn, C., Dalal, C., Cai, L., & Elowitz, M. (2015). Combinatorial gene regulation by modulation of relative pulse timing. Nature, 527(7576), 54. doi:10.1038/nature15710

Lou, C., Stanton, B., Chen, Y.-J., Munsky, B., & Voigt, C. (2012). Ribozyme-based insulator parts buffer synthetic circuits from genetic context. Nature Biotechnology, 30(11), 1137. doi:10.1038/nbt.2401

Lowe, D. (2018). Rise of the biologics. Med. Chem. Commun., 9, 10. doi: 10.1039/c7md90046e

Luo, Z., Zeng, W., Du, G., Chen, J., & Zhou, J. (2019). Enhanced pyruvate production in Candida glabrata by engineering ATP futile cycle system. ACS Synthetic Biology. doi:10.1021/acssynbio.8b00479

Maeda, Y., & Sano, M. (2006). Regulatory Dynamics of Synthetic Gene Networks with Positive Feedback. Journal of Molecular Biology, 359(4), 1107–1124. doi:10.1016/j.jmb.2006.03.064

Marangi, M., & Pistritto, G. (2018). Innovative Therapeutic Strategies for Cystic Fibrosis: Moving Forward to CRISPR Technique. Frontiers in pharmacology, 9, 396. doi:10.3389/fphar.2018.00396

Miliotou , A.N. & Papadopoulou, L.C. (2018). “CAR T-cell Therapy: A New Era in Cancer Immunotherapy”, Current Pharmaceutical Biotechnology, 19: 5. https://doi.org/10.2174/1389201019666180418095526 Min, Y.-L., Li, H., Rodriguez-Caycedo, C., Mireault, A., Huang, J., Shelton, J., … Olson, E. (2019). CRISPR-Cas9 corrects Duchenne muscular dystrophy exon 44 deletion mutations in mice and human cells. Science Advances, 5(3), eaav4324. doi:10.1126/sciadv.aav4324

Mundt, M., Anders, A., Murray, S., & Sourjik, V. (2018). A system for gene expression noise control in yeast. ACS Synthetic Biology. doi:10.1021/acssynbio.8b00279

Mutalik, V., Guimaraes, J., Cambray, G., Lam, C., Christoffersen, M., Mai, Q.-A., … Endy, D. (2013). Precise and reliable gene expression via standard transcription and translation initiation elements. Nature Methods, 10(4), nmeth.2404. doi:10.1038/nmeth.2404

Nelson, C., Wu, Y., Gemberling, M., Oliver, M., Waller, M., Bohning, J., … Gersbach, C. (2019). Long-term evaluation of AAV-CRISPR genome editing for Duchenne muscular dystrophy. Nature Medicine, 25(3), 427–432. doi:10.1038/s41591-019-0344-3

Nielsen, A., Der, B., Shin, J., Vaidyanathan, P., Paralanov, V., Strychalski, E., … Voigt, C. (2016). Genetic circuit design automation. Science, 352(6281), aac7341. doi:10.1126/science.aac7341

Peng, S., Stephan, R., Hummerjohann J. & Tasara, T. (2014). Evaluation of three reference genes of Escherichia coli for mRNA expression level normalization in view of salt and organic acid stress exposure in food. FEMS Microbiol Lett.; doi:10.1111/1574- 6968.12447.

Peters, M., Meinhold, P., Glieder, A., & Arnold, F. (2003). Regio- and Enantioselective Alkane Hydroxylation with Engineered Cytochromes P450 BM-3. Journal of the American Chemical Society, 125(44), 13442–13450. doi:10.1021/ja0303790

Purvis, J. E., & Lahav, G. (2013). Encoding and Decoding Cellular Information through Signaling Dynamics. Cell, 152(5), 945–956. doi:10.1016/j.cell.2013.02.005

Qi, L., Haurwitz, R., Shao, W., Doudna, J., & Arkin, A. (2012). RNA processing enables predictable programming of gene expression. Nature Biotechnology,30(10), 1002. doi:10.1038/nbt.2355

Ran, F., Cong, L., Yan, W., Scott, D., Gootenberg, J., Kriz, A., … Zhang, F. (2015). In vivo genome editing using Staphylococcus aureus Cas9. Nature, 520(7546), 186. doi:10.1038/nature14299

Rege, M., Kim, J., Valeri, J., Dunagin, M., Metzger, A., Gong, W., … Phillips-Cremins, J. (2018). LADL: Light-activated dynamic looping for endogenous gene expression control. bioRxiv, 349340. doi:10.1101/349340

Rogers, J., Guzman, C., Taylor, N., Raman, S., Anderson, K., & Church, G. (2015). Synthetic biosensors for precise gene control and real-time monitoring of metabolites. Nucleic Acids Research, 43(15), 7648–7660. doi:10.1093/nar/gkv616

Rosenfeld, N., Elowitz, M., & Alon, U. (2002). Negative Autoregulation Speeds the Response Times of Transcription Networks. Journal of Molecular Biology,323(5), 785– 793. doi:10.1016/S0022-2836(02)00994-4

Shalem, O., Sanjana, N. E., Hartenian, E., Shi, X., Scott, D. A., Mikkelson, T., … Zhang, F. (2013). Genome-scale CRISPR-Cas9 knockout screening in human cells. Science (New York, N.Y.), 343(6166), 84–87. doi:10.1126/science.1247005

Sheridan, C. (2018). Go-ahead for first in-body CRISPR medicine testing. Nature Biotechnology. doi:10.1038/d41587-018-00003-2 Shin, Y., Berry, J., Pannucci, N., Haataja, M., Toettcher, J., & Brangwynne, C. (2017). Spatiotemporal Control of Intracellular Phase Transitions Using Light-Activated optoDroplets. Cell, 168(1-2), 159–171.e14. doi:10.1016/j.cell.2016.11.054

Stricker, J., Cookson, S., Bennett, M., Mather, W., Tsimring, L., & Hasty, J. (2008). A fast, robust and tunable synthetic gene oscillator. Nature, 456(7221), 516. doi:10.1038/nature07389

Taylor, N., Garruss, A., Moretti, R., Chan, S., Arbing, M., Cascio, D., … Raman, S. (2015). Engineering an allosteric transcription factor to respond to new ligands.Nature Methods, 13(2), nmeth.3696. doi:10.1038/nmeth.3696

Teo, J., Woo, S., & Sarpeshkar, R. (2015). Synthetic Biology: A Unifying View and Review Using Analog Circuits. IEEE Transactions on Biomedical Circuits and Systems, 9(4), 453–474. doi:10.1109/TBCAS.2015.2461446

Wang, H., Yang, H., Shivalila, C., Dawlaty, M., Cheng, A., Zhang, F., & Jaenisch, R. (2013). One-Step Generation of Mice Carrying Mutations in Multiple Genes by CRISPR/Cas-Mediated Genome Engineering. Cell, 153(4), 910–918. doi:10.1016/j.cell.2013.04.025

Wannier, T., Kunjapur, A., Rice, D., McDonald, M., Desai, M., & Church, G. (2018). Adaptive evolution of genomically recoded Escherichia coli. Proceedings of the National Academy of Sciences, 115(12), 201715530. doi:10.1073/pnas.1715530115

Wong, T., Arnold, F., & Schwaneberg, U. (2004). Laboratory evolution of cytochrome P450 BM‐3 monooxygenase for organic cosolvents. Biotechnology and Bioengineering, 85(3), 351–358. doi:10.1002/bit.10896

Xue, S., Tian, S., Fujii, K., Kladwang, W., Das, R., & Barna, M. (2015). RNA regulons in Hox 5′ UTRs confer ribosome specificity to gene regulation. Nature, 517(7532), 33. doi:10.1038/nature14010

Zhao, E., Zhang, Y., Mehl, J., Park, H., Lalwani, M., Toettcher, J., & Avalos, J. (2018). Optogenetic regulation of engineered cellular metabolism for microbial chemical production. Nature, 555(7698), 683. doi:10.1038/nature26141

Protocol Attachment

SOCi Overview This protocol covers the creation of SOC (super optimal broth with catabolite repressionii) media. Makes 200mL.

Materials - Tryptone o Final concentration: 2% (20g/L) - Yeast Extract o Final concentration: 0.5% (5 g/L) - NaCl o Final concentration: 10 mM - KCl o Final concentration: 2.5 mM

- 1M MgCl2 o Final concentration: 10 mM

- 1M MgSO4 o Final concentration: 10 mM - Clean 500mL flask - Filter Sterilized 20% Glucose o Final concentration: 20 mM o Note that 20% glucose is ~1.11M - Millipore water - Sterile water

Procedure 1. In a flask, add 4g Tryptone, 1g Yeast extract, 0.117g NaCl and 0.037g KCl. Add Millipore water to 150mL. Swirl to mixiii 2. Autoclave.

3. In the dead air box, add 2mL 1M MgCl2, 2mL 1M MgSO4 and 3.6mL 20% Glucose.

i Author: Ethan M Jones ii i.e. glucose. This is because the presence of glucose shuts down/down regulates pathways associated with the consumption of non-glucose sugars (e.g. arabinose and lactose). This is due to the fact that glucose is a preferred energy source, as well as that most of these pathways are auto regulatory (i.e. they convert sugars to glucose, which then represses that pathway, causing the right amount of enzymes to be produced). iii There is technically a pH adjustment step to pH 6.8-7.0 (although Upsula iGEM 2011 say pH 7.5) using NaOH, however as SOC is unbuffered like LB, and since it gets used for such a short period of time (and should naturally be at a neutral pH), we don’t perform this step. 4. Add sterile water to 200mL 5. Aliquot in 50mL increments in a clear tube. Store at RT o Clearness is important for checking for contamination o Alternatively aliquot into sterilized 25mL square tubes from NEBiv

iv These should be autoclavable as they are polycarbonate, however they will decay over time. Qiagen Buffer PEi Overview This protocol covers the creation of the Qiagen PE wash buffer. Makes 50mL.

Materials - 100% (absolute) Ethanol o Final concentration 80% v/v - 1M Tris-HCl pH 7.5 o Final concentration 10mM o Buffer should be clear, orange or yellowish color is indicative of low quality Tris - 50mL Falcon Tube Procedure 1. Add 500µL of 1M Tris-HCl pH 7.5 to a 50mL falcon tube 2. Add 9.5mL of sterile water 3. Add 40mL of absolute ethanol 4. Invert to mix. Store at room temperature

i Author: Ethan M Jones DNA Ladderi Overview This protocol covers the creation of DNA Ladder working solution.

Materials - Labeled 1.7mL microcentrifuge tubes o Labels need to be say the DNA Ladder name - NFW - 6x Purple Loading Dye (NEB)ii - Appropriate DNA Ladder (typically Invitrogen 1kb+) o Make sure to note what concentration of DNA is meant for a single lane, as well as the concentration of the DNA. This is almost always .5µg per lane and .5µg/µL concentration, but it is good to check.

Procedure 1. Thaw DNA Ladder (you may hand thaw) 2. Add to the labeled tube the appropriate amount of loading dye, NFW and ladder to make the correct loading amount per lane 6µL. o In general (1µL 6x loading dye, 1µL ladder, 4µL water), so to make 1.2mL of ladder (200 lanes), add 200µL dye, 200µL DNA ladder, 800µL NFW. 3. Store in gel loading rack at 4Ciii.

i Author: Ethan M Jones ii NEB loading dye generally works better than most other loading dyes, but ultimately it isn’t a big deal. I’m not a fan of the “BlueJuice” dye that comes with 1kb+ ladder; it doesn’t load well, is too thin, and leaves a shadow when imaged. iii The ladder is a mix of double stranded DNA which is not actually going to be used in an application, as such it can be stored for months to years at 4C, although the stock should be stored at -20C Qiagen Buffer P2i Overview This protocol covers the creation of the Qiagen buffer P2 (lysis), for use in miniprep (alkaline lysis method).

Materials - 100mL Glass Bottle - SDS (Sodium dodecyl sulfate) - Millipore water - NaOH (Sodium Hydroxide) o NaOH is caustic, work in a chemical fume hood. Add (small amounts of) hydroxide to water, not the other way around. Be careful with aliquoting this solution, not all plastics are stable to NaOHii o The dissolution of NaOH in water is a strongly exothermic process, add NaOH in small amounts and consider performing on ice. o Molar Mass: 39.997

Procedure 1. The final concentrations for buffer P2 are 1% SDS and 200mM NaOH in water. o This protocol assumes that you do not have stock solutions of these chemicals (e.g. 10% SDS and 1M NaOH). If you have stock solutions, simply mix with Millipore water in the appropriate amounts. 2. For each 100mL of buffer P2, you will need .8g of NaOH and 1g of SDS. The following assumes you are making 100mL of solution, scale up or down as appropriate. 3. Weigh and measure out NaOH and SDS. 4. Add 80mL of Millipore water to the bottle. 5. In a chemical fume hood, add SDS. o Light heating may be needed 6. In small portions add NaOH. Be careful as bottle may be hot. 7. Once solids have dissolved, adjust volume to 100mL with Millipore water. 8. Store at room temperatureiii.

i Author: Ethan M Jones ii See https://www.calpaclab.com/chemical-compatibility-charts/ for details iii Note that if stored at lower temperatures, white bubbly solid may form. This is reversible, simply heat until dissolve, then store at appropriate temperature.

10X ETBRi Overview This protocol covers the creation of a 10x stock of ethidium bromide for the dropper bottle

Materials - 50mL Centrifuge tube - Ethidium Bromide Powder - Tin foil - Millipore water

Procedure 1. A 10x stock has a concentration of 5mg/mL (dropper working concentration 500µg/mL). 2. To a 50mL centrifuge tube, add 40mL Millipore water and 0.250g ethidium bromide. Vortex until dissolved. 3. Adjust volume to 50mL with Millipore water. Cover tube with foil and store at 4C.

i Author: Ethan M Jones 100% Glyceroli Overview This protocol covers the sterilization of pure glycerol. While glycerol is can be autoclaved, it ideally should not be autoclaved repeatedlyii. For usage in glycerol stocks however, it shouldn’t matter at alliii.

Materials - 100% glycerol - Glass bottle

Procedure 1. Fill glass bottle halfway with 100% glycerol. 2. Autoclave on liquid cycle. Store at room temperature

i Author: Ethan M Jones ii Potential production of acrolein. Probably not relevant though. iii The fact of the matter is most people autoclave their glycerol, and indeed many people autoclave media with glycerol in it with no ill effects. Honestly, for glycerol stocks, autoclaving might not even be needed (we never used to back in 2016). After all, what exactly is going to be growing in glycerol (it’s even synthesized from petrochemicals, so that’s one less contamination vector). 50% Glyceroli Overview This protocol covers the creation of a 50% v/v glycerol solution, which is typically used as a cryoprotectant for glycerol stocks at -20C .

Materials - 250mL glass bottle - 100% glycerol - Millipore water

Procedure 1. Add 75mL of glycerol to a glass bottle o Consider using a 50mL centrifuge tube 2. Add 75mL of Millipore water. 3. Mix. Autoclave on liquid cycle 4. Aliquot into 50mL centrifuge tubes 5. Store at room temperature

i Author: Ethan M Jones 87% Glyceroli Overview This protocol covers the creation of a 87% v/v glycerol solution, which is often used in place of 100% glycerol as it is significantly easier to work with (less viscous). This protocol covers making 50mL aliquots, which are used for making glycerol stocks for storage at -80C.

Materials - Sterile 100mL glass graduated cylinder - Parafilm - Sterile 100% glycerol - Sterile water

Procedure 1. In the hood, add 13mL of sterile water, then add 100% glycerol to 100mL 2. Cover with parafilm and mix well. 3. Aliquot into 50mL centrifuge tubes 4. Store at room temperature

i Author: Ethan M Jones LBi Overview This protocol covers the creation of 1 liter of Miller Lysogeny Broth (LB). A double concentrated stock is made, sterilized via autoclave and then diluted to final working concentration with sterile water.

Materials - 1L Autoclavable glass bottle - Millipore Water

- Sterile H2O Either - 25g Miller LB premix o Final concentration: 25g/L Or - NaCl (Sodium chloride) o Final concentration: 10 g /Lii - Tryptoneiii o Final concentration: 10g/L - Yeast extractiv o Final concentration: 5g/L Procedure 1. Weigh out either 25g of LB premix or 10g NaCl, 10g Tryptone, 5g Yeast extract. o Note that LB premix, Tryptone and Yeast Extract have a tendency to get everywhere. After you measure and pour ingredients into the bottle, make sure to clean the benches and the balance extremely well. 2. Pour into a labeled 1L glass bottle.

o Label should include LB, the date, initials and a checkbox for ddH2O 3. Add Millipore water to 500mL, autoclave on liquid cycle. o Note that LB should be autoclaved in fairly short order. If left out at room temperature, “things” will begin to grow.v

i Author: Ethan M Jones ii This makes it Miller Broth. We have found that salt really should makes much of a difference as to performance, especially since we don’t use LB for experiments. iii Tryptone is a mixture of short peptides and amino acids, and is made by digesting casein with the protease trypsin iv Dead, dried yeast whose cell walls have been broken down. Contains a large number of micronutrients and other miscellaneous organic molecules as well as amino acids/peptides. v Putting it at 4C overnight should be able to stave it off somewhat, but the clock is ticking. 4. Clean benches and balance thoroughly.

vi 5. In the hood using sterile technique, add Sterile H2O to 1L. Check off checkbox.

vi Some people would say that LB should be pHed to 7.2, but that is never a relevant issue. 20% glucosei Overview This protocol covers the creation of a 20% w/v glucose solution, which is typically used as a media supplement. 20% w/v glucose is ~1.11 M. Note that this solution must be filter sterilized, as glucose cannot be autoclavedii.

Materials - 1L glass bottle. - 1L graduated cylinder - Glass beaker - Glucose (Dextrose) - Millipore water - Stir bar - 0.22µm vacuum filter

Procedure 1. Weigh out 200g of glucose 2. Add ~800mL of Millipore water to the glass beaker. 3. Dissolve glucose with stirring, adding portions of glucoseiii 4. Stir until dissolved. Heat very gently if needed. 5. Transfer to graduated cylinder, adjust volume to 1L with Millipore water. 6. Vacuum filter into a sterile labeled glass bottle. 7. Ideally, aliquot in 250mL bottles. 8. Store at room temperature

i Author: Ethan M Jones ii This is due to caramelization. Medias which contain glucose are similarly unable to be autoclaved. From personal experience I can say that the solution does smell like caramel iii If you don’t do this, you can end up with rock candy. (i.e. really hard to dissolve) 10% Casamino Acidsi Overview This protocol covers the creation of a 10% (w/v) stock solution of Casamino Acids.

Materials - Clean 1L graduated cylinder - 1L glass beaker. - Millipore water - Casamino acids - 0.22µm bottle top filter - Sterile 1L glass bottle - ~25 50mL centrifuge tubes labeled 10% Casamino acids

Procedure 1. Weigh out 100g of casamino acids into a 1L glass beaker o Make sure to clean balance well after usage 2. Add ~850mL of Millipore water to a glass beaker, along with a stir bar. o Recommended to add water slowly at first 3. Add the casamino acids (consider using a funnel). Add Millipore water to ~950mL 4. Dissolve casamino acids with gentle heating/stirring. 5. Cool (if heated). Transfer to 1L graduated cylinder. Adjust volume with Millipore water to 1L. 6. Filter sterilize with a 0.22µm bottle top filter. 7. Using pipette controller, aliquot 40miiL into each centrifuge tubeiii.

i Author: Ethan M Jones ii This is the amount needed for 1L of M9 iii Added since large bottles tended to get contaminated due to poor technique after a few uses (usually by fuzzy balls of fungus) 1.25x loading dyei Overview This protocol covers the creation of 1.25x loading dye. 1.25x (sometimes called 1.2x) loading dye is used to prevent the tedium of mixing loading dye with water for every piece of DNA run on a gel. Since we almost run 1µl of DNA (usually from a PCR), we prepare a loading dye solution that works well with that.ii

Materials - 6x Purple loading dye (NEB) o While other loading dyes exist, we use (and have vast amounts ofiii) purple loading dye from NEB.iv o 6x Loading dye is stored either in the gel rack at 4C or at -20C. o Note: Do not use the loading dye with no SDS, that is needed for gel extractions. - Nuclease Free Water (NFW) - Labeled 1.7mL tube(s) o Only the top need be labeled. Label simply needs to read 1.25x Gel Loading Dye. If so desired you may add this on the side along with the date. Procedure 1. Pipette a 1:4 ratio of 6x loading dye to NFW into a labeled tube o (e.g). 300µl 6x loading dye,1200µl NFW) 2. Store in fridge on gel loading rack. o Diluted loading dye is stable at 4C

i Author: Ethan M Jones ii For colony PCRs where 2µl of PCR product is run, it is not a big deal. The difference is minor and it works fine. iii This is because they send it with basically every enzyme the deals with DNA. iv We like this dye because it does not leave a shadow on the gel, and doesn’t disperse into the solution during loading. Dropper Bottle ETBRi Overview This protocol covers refilling the ETBR dropper bottle with ethidium bromide solution, this allows us to use the extremely cheap ETBR powder (Saha lab has 10x lifetime supply) rather than paying exorbitantly for a premade dropper bottle). The appropriate concentration of ETBR in an agarose gel is ~0.5µg/mLii. Given that the dropper bottle we useiii has a drop volume of .05mL, and that our gels are 50mL in size, the appropriate concentration for the dropper bottle solution is 500µg/mL.

Materials - Dropper bottle - 10x Stock ETBR (5mg/mL) - Millipore water

Procedure 1. Add 500µl of 10x Stock ETBR to dropper bottle. Then add 4.5mL of Millipore water. Give at least 1 minute to mix. 2. Store at room temperatureiv

i Author: Ethan M Jones ii addgene.org/protocols/gel-electrophoresis/ (this value is actually fairly flexible, I know that Saha lab likes to use like 3x as much) iii Also dropper bottles more generally. iv ETBR is really quite heat stable, and besides if the solution did happen to get weaker (has never occurred), just use more or make a fresh bottle. 20% Glyceroli Overview This protocol covers the creation of a 20% v/v glycerol solution, which is typically used as a media supplementii.

Materials - Parafilm - 1L graduated cylinder - 100% glycerol - Millipore water - Sterile 1L bottle - 4 sterile 250mL bottles - 0.22µm bottletop filteriii

Procedure 1. Add 200mL of glycerol to a graduated cylinder o Consider using a 50mL centrifuge tube 2. Add 800mL of Millipore water. 3. Cover with parafilm, mix well. 4. Filter sterilize. Aliquot into 250mL bottles 5. Store at room temperature

i Author: Ethan M Jones ii Carbon source. Particularly when glucose might have unwanted effects (e.g. arabinose induction) iii We filter our media glycerol supplements just in case autoclaving might matter (probably doesn’t) Annealing 3G Adaptersi Overview This protocol covers the resuspension and subsequent annealing of dried (lyophilized) oligos that are to be used as 3G adapters. Single stranded oligos are resuspended, and then annealed to each other before being diluted. In this protocol un-annealed stock (100µM), duplex stock (5µM) and duplex working (50nM) solutions are made. This protocol also serves as a general purpose protocol for the annealing of complementary oligos.

Background During the Golden Gate portion of 3G assembly, double stranded UNS adapters are used to add PCR amplification and Gibson Assembly launching pads onto transcriptional units. These double stranded adapters are ~50bp long and contain BsaI cut sites, sticky ends and a UNS sequence. Since these sequences are short, they can be purchased as two complementary single stranded oligos and then annealed together.

Materials - Lyophilized Oligos (IDT) o Each oligo should have a corresponding specification sheet, which should be stored in the specification sheet drawer. - Labeled 1.7mL centrifuge tubes (2 per sample) o Tubes should be labeled with adapter name and 5µM, dates and initials should be included on the side. A second tube should be labeled with 50nM. - 0.2mL Tubes - IDT Duplex Buffer o Composed of: 30 mM HEPES, pH 7.5; 100 mM potassium acetate o Stored at 4C o Use the falcon tube aliquot, not the stock! - Nuclease Free Water (NFW)

Procedure 1. Briefly spin down oligo tubes. o Use green benchtop centrifuge 2. Determine correct amount of IDT duplex buffer to add to each tube for a 100µM solution. This number can be found on the specification sheet. Alternatively, it can be determined by multiplying the number of nanomoles of DNA provided by

i Author: Ethan M Jones 10 and then using that many µl of duplex buffer. Remember that each tube will have a different amount of duplex buffer to add. 3. Add correct amount of duplex buffer to each tube. Heat at 55C for 2 minutes. Vortex. Briefly spin down. 4. Add 5µL of each the two corresponding oligos to a .2mL tube. o Each oligo is now at a concentration of 50µM 5. Then add 90µL of IDT duplex buffer. o Each oligo is now at 5µM concentration 6. Program thermocycler to heat at 94C for 2 minutes, and then descend by 1C per minute until 20C is reached. This can easily be done using the gradient thermocycler’s “delta” feature. 7. Once annealing is finished, transfer 5µM duplex to duplex stock tube 8. Next make a 50nM working solution by 1:100 diluting duplex stock in NFW (eg. 1µL duplex to 99µL NFW). 9. Store stock and working duplexes in duplex box. Log on inventory sheet.

Resuspension of Primersi Overview This protocol covers resuspension of dried (lyophilized) oligos that are to be used as primers. In this protocol both a stock (100µM) and working (10µM) solutions are made.

Background Primers are short (typically less than <60bpii) chemically synthesized DNA oligonucleotides used in PCR and related applicationsiii. Typically oligos are purchased from a supplieriv and shipped in lyophilized formv, and are then resuspended for use. Since relatively little primer is needed per reaction, a stock solution and a working aliquot is usually made.

Equipment: - Picofuge - Heatblock

Procedure 1. Briefly spin down oligo tubesvii. o Use green benchtop centrifuge

i Author: Ethan M Jones ii Normally ~17-30bp for standard PCR, but our overhangs for Gibson make our typical primers longer iii E.g. Reverse Transcription iv Typically IDT or Twist, occasionally someone such as Genscript or any of the numerous smaller players v DNA is much more stable in lyophilized form (since water is needed for hydrolysis), IDT states their lyophilized oligos should be fine at 25C for at least 6 months (and I’d imagine a great deal longer). vi Other labs might use TE buffer, which is better for long term storage. As we’ve never had a problem with degradation, and switching to TE could conceivably cause minor problems downstream (from EDTA), we use NFW. Additionally, switching now would create an inconsistency. vii This is to ensure that all of the oligo is at the bottom of the tube. 2. Determine correct amount of NFW to add to each tube for a 100µM solution. This number can be found on the specification sheet. Alternatively, it can be determined by multiplying the number of nanomoles of DNA provided by 10 and then using that many µl of NFW. Remember that each tube will have a different amount of NFW to add. 3. Add correct amount of NFW to each tube. Heat at 55C for 2 minutes. Vortex. Briefly spin down. 4. Prepare a 10µM working stock by adding a NFW and 100µM primer at a ratio of 9:1 (9µl NFW for every 1µl primer). Typically we make a 100µl worth of working solution. 5. Store primers in appropriate boxes in -20. Log their locations on google drive. Update the location column on appropriate primer excel sheet in the inventory folder on Dropbox.

Transformationi Overview This protocol covers transformation of chemically competent E.coli cells from NEB with plasmids from either miniprep or Gibson Assembly. See additional information for instructions about transforming homemade competent cells.

Background Often, it is desired to place a given plasmid in E. coli cells. The two main reasons for giving E. coli plasmids are to perform experiments (placing circuits in E. coli for measurement)ii, or for cloning (obtaining more of that plasmid). Competent cells, cells which are capable of taking up plasmids come in two main forms, electrocompetent and chemically competent. Electrocompetent cells require more equipment and are somewhat more involved, but typically have better transformation efficiencies (number of successful transformants compared to DNA input), while chemically competent cells are easier and quicker to make and useiii. Although it is not difficult to make competent cells of either type, due to (fairly) cheap high quality chemically competent cells being commercially available, many labs no longer make their owniv.

In general we use chemically competent cells from NEB, although we will transform homegrown cells when we know the transformation is likely to work (e.g. we have a confirmed miniprep and we want more of it).

Materials - Chemically competent cells from NEBv o Cells are stored at -80C and should be always be kept on ice. Note that once cells are thawed (removed from -80C), they cannot be refrozen and used again.vi o Tubes contain at least 50µl total volume. The minimum volume per transformation is 10µl. o We use 3 different cell types., 5-alpha, 10-Beta, BL21. 5-alpha is generally used for all cloning, while the latter two are used for experiments.vii - Temporary 1.7mL centrifuge tubes (1 per sample) i Author: Ethan M Jones ii There are other reasons why one might put a plasmid that expresses genes in a cell (e.g. protein purification). iii There are a number of seminal papers on plasmid transformation from the 1970s-1980s from which most modern methods are derived. If you ever wonder why specific labs use specific protocols, you can find out by finding out which paper their transformation method descends from. iv The real main benefit to using commercial cells (besides ease of use), is that it eliminates a potential error source. Commercial cells will always work consistently well (likely better than any homemade ones), as such as long as you follow a consistent protocol, you can be assured that failed transformations are not because the competent cells go bad. v Theoretically most of this protocol should work for any chemically competent cells. vi Technically they can be, but efficiency will be greatly reduced. (Hence it would not be worthwhile). vii 5-alpha is RecA deficient, which aids the maintenance of plasmids (recombination = bad). 10-Beta is incapable of metabolizing arabinose, which is useful when using the pBad promoter. 10-Beta has the RecA1 mutation, which reduces recombination. It is occasionally used as a cloning strain, particularly for large (>10 kb) plasmids. BL21 is a protein expression strain often used for creation and purification of protein, as such it is deficient in OmpT and Lon (proteases). o Each tube should be labeled with a key (typically numerical) that corresponds to a specific transformation - Ice bucket (full) - Plasmid(s) for transformation o Gibson Assembly reactions can be directly transformed while plasmid minipreps should be diluted to ~1-2ng/1nM concentration.viii - SOC/10-Beta media o 10-Beta media should only be used 10-Beta cells. SOC should not be used for 10-Beta cellsix

o SOC contains: 2% w/v tryptone, .5% w/v yeast extract, 10mM NaCl, 10mM MgCl2, 10mM x MgSO4, 20mM glucose - Antibiotic resistant plates o Plates should obviously have the appropriate antibiotic(s) for the plasmids you are working with. Plates should be pre-warmed (lid side down) before plating, and before labeling (hard to label a foggy plate).

Procedure 1. Remove cells from -80C and place onto ice. Wait ~10 minutes for cells to thaw. o Note that you should never vortex or centrifuge these cells at any point. o Cells should be kept on ice unless otherwise indicated 2. Add appropriate amount of cells to each keyed temporary transformation tube (minimum 10µl per tube). o If you are transforming less than 5 plasmids per tube, use slightly more cells. o For homemade competent cells, use 50µL per transformation o Tubes should be placed on ice before aliquoting cells 3. For each 50µL of cells add 2µl of each plasmid (diluted as above) or Gibson Assembly product to the appropriate tube. (i.e. add .5µL Gibson for 12.5µL of cells). o This should be done on ice 4. Flick tube 2-3 times to mix. Cells will be on the sides of tube, do not be alarmed. Do not centrifuge.

viii Efficiency is highest in the 1pg-1ng range (per 50µL cells), however the largest numbers of colonies will be gained by using 10ng per transformation (via NEB). Consider however that these statistics are for plasmids of a certain size, one would imagine that mols of DNA would be more important. See additional info for graph. ix Strictly speaking 10-Beta cells will probably be fine in SOC. The same may not be true for the reverse. Also our transformation efficiencies are usually so high that a minor decrease would not prove problematic. Do not take this as an invitation to use the wrong media. I think NEB has info about this on their website. x SOC (Super Optimal broth with Catabolite repression) is a derivation of the rich media SOB (Super Optimal Broth), which contains the same components as SOB but with the addition of glucose. 5. Preheat the heat block to 42C. Remember to both turn heat block on and to turn on heating element. 6. Allow cells to incubate on ice for 30 minutesxi 7. Heat shock (put in heat block) cells for exactlyxii 30 seconds (for 10-Beta, 5- alpha cells and JS006 ) or exactly 10 seconds (BL21 cells).xiii 8. Immediately return cells to ice. Incubate 5 minutes 9. Visually inspect SOC bottle to ensure it is not contaminated. Solution should be transparent, with no floating particles or cloudinessxiv. 10. Add SOC (5-alpha and BL21 cells) or 10-Beta media (10-Beta cells) to tubes. For each 1µl of cells used, you should use 19µl of media (e.g. 10µl of 5-alpha cells, 190µl SOC). 11. Place tubes in shaking incubator (250 RPM, 37C) for 1 hour (non kanamycin antibiotics) or 2 hours (kanamycin)xv 12. At this point you should pre-warm and label your plates. 13. Remove tubes from incubator. 14. Invert each tube at least 6 times and add an appropriate amount (50µl for Gibson Assembly, 100µL for 3G Assembly, 30µl for experiment plates, 100µL for homegrown competent cells) to each plate. 15. Add ~3 sterilized glass beads to each plate and shake side to side (rotate 90 degrees after a few shakes). 16. Remove glass beads into used glass bead container. Place plates (lid side down) in incubator at 37C. Colonies should form within 16-18 hours.

xi Going slightly longer (~10 minutes) than 30 minutes will not harm the cells. See additional info section. xii Cells cannot actually tell time. 29 seconds is fine, as is 31 seconds. See additional information for graph. xiii An “easy” way to remember this is that 10-Beta cells get 30 seconds and BL21 cells get 10 seconds. /s/ xiv This step is courtesy of AJO 2018 xv I’m honestly not sure this is needed, or even why it’s needed. I suppose for low efficiency transformations it will be worthwhile

Additional Information For homemade competent cells, follow the protocol as above, using a single tube (50µL of cells) for each transformation (950µL SOC, 30 second heat shock).

Effect of DNA incubation time on NEB 5-alpha competent E.coli transformation efficiency (via NEB):

DNA Effects on Transformation Efficiency and Colony Output (via NEB):

Effect of heat shock time on NEB 5-alpha competent E.coli transformation efficiency (via NEB): Plasmid Miniprepi Overview This protocol covers the isolation of plasmid DNA from E. coli via miniprep (alkaline lysis) using Qiagen reagents. This protocol also briefly covers the creation of a glycerol stock for preservation of plasmid containing bacteria. Note that this protocol does not include appropriate steps for the preparation of plasmids for use with eukaryotic cell cultureii (i.e. transfections, particularly in mammalian cell lines). For information on how to accommodate prepare plasmids for use in eukaryotic cell culture, please see Qiagen’s manualiii.

Background One of the foundational methods of molecular biology is plasmid preparation, the extraction and purification of plasmid DNA from bacteria. Plasmid preparation is so ubiquitous in molecular biology because it enables the large scale amplification of DNA in a format that is also amenable to bacterial expressioniv. Plasmid preparations can be done at different scalesv, the most common of which is the minipreparation or miniprep, which is most commonly used as it is high speed, inexpensive and usually sufficient for most purposesvi.

While there are a few different ways to perform minipreps the fundamental procedure always involves growing up a small (3-5mL) volume of bacterial cells (typically E. coli), lysingvii those cells (releasing plasmid and ideally trapping genomic DNA), and then performing one or more isolation and/or purification stepsviii. The most common form of miniprep performed today is a column based purification using an alkaline lysis method. In this method, E. coli are pelleted through centrifugation, resuspended and subject to alkaline conditionsix. These conditions cause the lysis of the cells, and a neutralization buffer is added to restore pH. This neutralization buffer typically contains a guanidine based denaturing agent, which serves to denature proteins. This method of lysis typically allows the smaller plasmid molecules to diffuse freely into solution while trapping genomic DNA in the cellular debrisx. The lysed solution is then pelleted, and

i Author: Ethan M Jones ii Additional steps must be taken as animal cells are extremely sensitive to endotoxins (i.e. toxins present inside bacterial cells, typically lipopolysaccharides [LPS], a component of gram negative bacterial cell walls) iii The only requirement is the addition of an additional denaturing step to remove potential endotoxins. This generally lowers purity and yield (as well is a waste of time) and so in the absence of a need is not performed. iv It was also discovered very early on in the history of molecular biology, and was amenable to the methods available before PCR. v Mini, midi, maxi, mega, giga vi For many methods you just don’t need that much DNA. For transfections and probe synthesis, midi preps are more common, but even then the cultures used for midi preps are typically derivided from minipreps. vii Usually via alkaline lysis, but occasionally using lysozyme, phenol/chloroform, boiling or mechanical methods like sonication. viii Typically a phenol chloroform based isolation (often with isoamyl alcohol) followed by ethanol purification steps. ix Along with a surfactant x This is why vortexing and shaking is to be avoided, as shearing of genomic DNA leads to small fragments which are able to escape the debris and contaminate the prep. the supernatant is then run through a column based DNA purification (see PCR purification protocol for information on mechanics).

Materials - E. coli culture grown overnight (14-16 hours) containing plasmid to be isolated. o Culture should be in late exponential growth phase (turbid but not overly so). Overlong growth of culture will lead to lower yields and lower quality plasmid. Shortened growth of culture leads to lower yield o The maximum volume for a single miniprep column is 5mL of culture. Note that higher culture volumes are associated with higher yield.xi o Standard culture volume is 4mL (3mL for miniprep, .5mL for glycerol stock) - Temporary 1.7mL centrifuge tubes (1 per sample) o Each tube should be labeled with a key (typically numerical) that corresponds to a specific culture tube. This key should be logged on your benchling. - Econospin Mini Spin Column (1 per sample) o Make sure you use the correct column (30-40µg binding capacity). Do not use the “min- elute columns”. Correct columns are blue and stored at room temperature. o Label the column with the same numerical keys as your temporary 1.7mL centrifuge tubes. Note, label the column (top part), not the collection tube which will be disposed of. - Final 1.7mL centrifuge tubes (2 per sample) o 1 tube is need for final miniprep tube and another for the glycerol stock o Each tube should be labeled with the construct name, plasmid backbone, the miniprep/glycerol stock number (abbreviate mp/gs. [e.g. mp#1]), the date and initials. o Initials and date only need to be located on the sides of tubes, not the top. o Glycerol stocks and miniprep numbers should correspond (i.e. mp#1 should come from the same culture as gs#1). - 50% Glycerol

o 50:50 volume to volume solution of Sterile H2O and Glycerol o Used to help cryopreservation of bacteria - Buffer P1 (Resuspension Buffer) o Buffer stored at 4C (fridge) o Comprised of: 50mM Tris-HCl pH 8.0, 10mM EDTA, 100µg RnaseA o Used to resuspend bacteria in buffered solution. RnaseA breaks down RNA after cells are lysed.xii - Buffer P2 (Lysis buffer) o Comprised of: 200mM NaOH, 1% SDS o Used to break down cell walls/membranes and denature proteins - Buffer N3 o Comprised of: 4.2 M Gu-HCl, 0.9 M Potassium acetate, pH 4.8

xi One would expect the 2nd derivative of plasmid per unit volume of culture to be negative xii While for our work we would consider RNA a fairly benign/minor contaminant, it represents a major contaminant for researchers working with eukaryotic transfections, or more commonly with in vitro transcriptions. o Neutralizes the base from the lysis buffer. - Buffer PE (Wash buffer) o Comprised of:xiii 10 mM Tris-HCl pH 7.5, 80% ethanol. Used to wash away proteins, salts, and various other contaminants. - Buffer EB (Elution Buffer) o Comprised of: 10 mM Tric-Cl pH 8.5 o Used to elute DNA from column

Equipment - Microcentrifuge

Procedure 1. Transfer 1.5mL of each culture to the appropriate temporary tube (P1000 set to 750µl recommended). o Vortex or pipette up and down culture before transfer, otherwise bacteria will clump on the bottom. 2. Pellet cells by centrifugation in microcentrifuge at 6000 RPMxiv. 3. Discard (pour out) liquid. o Do not bang or shake tube vigorously as this can lead to loss of bacterial pellet. o Try to remove as much liquid as can be done in a reasonably quickly manner. You will repeat this step so it does not matter if every last drop is gone. 4. Repeat steps 1-3 until desired volume of culture is spun down. o Recommended minimum volume: 3mL o Maximum value: 5mL 5. Remove excess liquid by blotting the tube on a paper towel. Be careful to not dislodge the pellet. o Note that the paper towel should not go inside the tube/touch the pellet. o Get as much as can be reasonably removed. Excess liquid will reduce yield/purity, but there is a limit to what can reasonably be removed without pipetting.

xiii Qiagen has not revealed the exact recipe, this is a close approximation. Source: https://openwetware.org/wiki/Qiagen_Buffers xiv Ideally these should be given in xG, but honestly, the measurements do not need to be that precise so variation between different microcentrifuges should be unimportant. 6. Resuspend pellet in 250µl of cold buffer P1. (To resuspend either vortex tube or pipette up and down on pellet until no solids remain). Pipetting is recommended. 7. Add 250µl of buffer P2. Gently invert at least 6 times to mix. Do not shake, do not vortex. 8. Incubate for 1-4 minutes. o Note: Incubation should not be allowed to continue for longer than 4 minutes (including time taken to neutralize). The true upper limit of incubation time is 5 minutes. Do not allow incubation to proceed longer than this as it will damage DNA. 9. Neutralize lysis buffer by adding 350µl of N3 to tube and gently inverting at least 6 times to mix. A white precipitate should form. Do not shake, do not vortex. 10. Pellet cell debris by centrifugation for 5-10 minutes at 13,300 rpm. o Make sure to face hinges of tubes outwards so debris collects on the same area of the tube 11. Aspirate liquid (being sure to avoid collection of precipitate) and apply to appropriate spin column. Spin for 1 minute at 6,000 rpm. Throw away temporary tube. o Maximum volume that can be used at once is 800µl. In the event there is too much liquid simply spin multiple times. o Adding precipitate to columns will cause reduced purity/protein contamination. Avoiding dispensing pipette in tube. It is expected that there will be a small volume that is impossible to aspirate without precipitate. Leave that volume behind. 12. Discard flow through (dump out fluid from lower collection tube, then replace the upper column). o Note: be careful to not bring the bottom of the column in contact with the liquid. 13. Add 800µl of buffer PE. Allow columns to stand for 1 minute. Spin for 1 minute at 13,300 rpm. Discard flow through. 14. Dry column by spinning empty column for 1 minute at 13,300 rpm. 15. Remove upper column from collection tube and transfer to the appropriate final miniprep tube. Throw away collection tube. 16. For high copy plasmids (e.g. 1C3), apply 25µl of Buffer EBxv to the column membrane. For low copy plasmids add 15µL of Buffer EB to the column membrane. Do not pierce membrane

xv Pre-warming elution buffer to 50C is supposed to increase yield, especially of large plasmids. o Note: The volume of elution buffer can be adjusted up or down. Higher elution volumes lead to higher total yields (ng of DNA) but lower concentrations (ng/µL). Lower volumes yield lower total yields but higher concentrations. The minimum total volume (i.e. elution buffer used in steps 16-18) is 30µl. 17. Let incubate 1 minute. Spin at 13,300 RPM. o For improved yield, let sit for 4 minutes instead. 18. Repeat steps 17 and 18.xvi 19. Throw away column. Measure concentration and purity via nanodrop. Record purities in log. 20. Store minipreps in the awaiting confirmation box at -20 and record their names, locations, concentrations and purities on the awaiting confirmation spreadsheet on google drive. 21. Create a glycerol stock by adding 500µl of culture to 500µl of 50% glycerol in your final glycerol stock tube. Invert tubes to mix. 22. Store in the awaiting confirmation box at -80C and record initials, dates and locations on the awaiting confirmation spreadsheet on google drive. 23. It is recommended to clean your pipette by wiping with 70% ethanol.

xvi Alternatively spin through the full volume and the elute again using the EB buffer just spun through (as per what is done in PCR purification). This method might lead to slightly higher yields, but is somewhat more labor intensive (in my opinion). Either works fine. Gene Block Resuspensioni Overview This protocol covers resuspension of IDT gBlocks to the working concentration of 0.1 pmol/µL.

Background Gene fragments also called gBlocks (IDT), are synthesized linear dsDNA sequences which can be used for a number of applications. Gene fragments are useful in that they enable the creation of complex sequences without PCR, as well as in that they allow for a gene or genetic circuit to be designed and cloned without access to template sequences. The largest limiting factor for gene fragments is their cost, as well as their limitations on sequence lengthii (though for cloning, multiple gene fragments may be used instead of a single large one). Gene fragments in tubes are typically delivered dried, and thus must be resuspended. This protocol assumes the use of IDT gBlocks, but theoretically any gene fragmentiii should use a similar protocol. The resuspension concentration of 0.1 pmol/µL was chosen due to its utility in Gibson Assemblyiv. Note that this concentration is the same as a concentration of 0.1µM, which could be useful if you intend to use your gene fragment for a different application (e.g. 3G assembly).

Note that IDT’s gBlocks can be amplified via PCR, though it is important that a high fidelity polymerase such as Q5 is used.

Materials - gBlocks Gene Fragment o For each gBlock, note the number of ng delivered as well as the number of fmol/ng o For reference, there are 1000 fmols/1 pmol - Nuclease Free Water (NFW)

Equipment - Picofuge o Note that IDT gBlocks come in 2mL capped tubes, which are not compatible with all picofuges - Vortexer

Procedure

i Author: Ethan M Jones ii Typically around 3kb max length iii The next most commonly used company for gene fragments is Twist Bioscience, though there are many other companies that make gene fragments. E.g. Thermofisher, Genscript, Genewhiz and many many more. iv Since we typically think in pmols for Gibson Assemblies. 1. Briefly spin gBlock in green benchtop picofuge to ensure that the DNA is at the bottom of the tube. 2. Add the appropriate amount of (µL) of NFW to each gBlock

-./0 ["# %&'()&*&%]∗[ ] o Amount of NFW in µL should be 12 344 3. Vortex for 0.5 seconds. Spin down. 4. Store in gBlocks box at -20C Qiagen PCR Purificationi Overview This protocol covers column based purification of PCR products, using Qiagen reagents and Epoch Life sciences columns.

Background PCR purification is used to isolate relatively small (>10kb) fragments of dsDNA from the mixture of proteins and nucleotides left over after PCR. In brief, a high salt chaotropic binding buffer (typically composed of salts of guanidine) serves to denature proteins (including DNA binding proteins) and disrupt (weaken) the interface between water and DNA. These two interactions combined with the salts serve to enhance adsorption (binding) of the DNA to the silica membraneii. The column is then washed in a slightly acid salt solution containing ethanoliii, and then the DNA is eluted in a basic solution.iv

Materials - PCR/DpnI product o Depending on the volume, temporary tubes may be needed. - Labeled 1.7mL centrifuge tubes (1 per sample) o Tubes should be labeled with a key, initials and date. Unless PCR product is to be preserved in which case it should be fully labeled with primers and template used. - Keyed blue EconoSpin spin column (with lid)v - Buffer PB o Guanidine and isopropanol-based binding buffer o Composed of: 5 M Gu-HCl, 30% Isopropanol - Buffer PE (Wash buffer) o Used to wash salts and other non DNA from the column o Comprised of:vi 10 mM Tris-HCl pH 7.5, 80% ethanol. Used to wash away proteins, salts, and various other contaminants.

i Author: Ethan M Jones ii Via Wikipedia: “The mechanism behind DNA adsorption onto silica is not fully understood… A further explanation of how DNA binds to silica is based on the action of guanidium HCl (GuHCl), which acts as a chaotrope. A chaotrope denatures biomolecules by disrupting the shell of hydration around them. This allows positively charged ions to form a salt bridge between the negatively charged silica and the negatively charged DNA backbone in high salt concentration. The DNA can then be washed with high salt and ethanol, and ultimately eluted with low salt” iii Presumably this is somewhat analogous to a 70% ethanol wash in ethanol based precipitation. iv From my reading it seems that most of the talk about salts might be nonsense. See openwetware for more info. For sure pH matters though, as does having the appropriate level of salts (I’m just not sure about how/why they matter). v Based on their higher yield in gel extraction we’ve switched to using the standard (non min-elute) DNA spin columns. In general the min-elute columns have not performed well (and have lower DNA binding capacity), and are only really usable for extremely small volume elutions (~6µL), if that. vi Qiagen has not revealed the exact recipe, this is a close approximation. Source: https://openwetware.org/wiki/Qiagen_Buffers - Buffer EB (Elution Buffer) o Used to elute DNA from column o Comprised of: 10 mM Tric-HCl pH 8.5

Equipment - Heatblock - Microcentrifuge

Procedure 1. Add Buffer PB to DNA in a 5:1 ratio of PB to sample (e.g. add 135µl Binding buffer to 27µl DNA).vii o PCR tubes can hold 200µl of total volume. If the total volume after adding buffer will exceed 200µl, transfer DNA to a keyed temporary 1.7µl tube then add buffer. 2. Mix either by pipetting up and down or by flicking and spinning. 3. Apply each sample to appropriate spin column. Close cap and centrifuge at 6000 rpm for 1 minuteviii. 4. Discard flow through (dump out fluid from lower collection tube, then replace the upper column). 5. Add 750µL of Buffer PE. Let stand for 1 minute. 6. Centrifuge at 13,300 rpm for 1 minute. Discard flow through. o It is important to discard flow through before performing dry spin. 7. Perform dry spin by centrifuging again at 13,300 rpm for 1 minute. 8. (Optional): Preheat Buffer EB to 55Cix 9. Remove top column and place into appropriate final 1.7mL tube. Throw away collection tube o Do not allow column to make contact with any liquid in collection tube. If contact is made, repeat dry spin. 10. Elute by adding between 6-20µl Buffer EB to the column membrane (typically 12- 15µl used).

vii Buffer PB from Qiagen comes with an optional pH indicator, which we do not use (or include in our homemade buffer). This is because out of the thousands of PCR purifications we’ve performed, we’ve never once had to add Sodium acetate (pH 5.0) and because it is possible that the indicator might have downstream effects (Qiagen says as much for “sensitive applications”). viii Anecdotal evidence is given on openwetware that this reduction in speed increases yield (due to longer DNA binding time) ix Supposedly increases yield, especially for large >5kb DNA. o Ensure that Buffer EB is applied to membrane, not sides of column. o Greater volumes lead to greater yields of DNA, but lower concentration. 11. Let stand 4 minutesx. Spin at 13,300 RPM for 1 minute. 12. (Optional): Repeat step 10 and 11, using eluate (liquid that just came out of column), in place of elution buffer. Let stand for 1 minute instead of 2. o This step increases yield and should be performed if yields have been low, or if a large amount of DNA is needed (e.g. same backbone being used for a variety of Gibson reactions). 13. Discard column. Measure and record concentration using nanodrop. Store at 4C short term (>1 week), or at -20C long term.

x Adaption from original 1 minute elution, which supposedly increases yield. Abbreviated Protocol 1. Add 5:1 ratio of buffer PB to sample 2. Load sample onto column and centrifuge at 6,000 RPM 3. Warm buffer EB to 50C 4. Discard flow through 5. Add 750µL buffer PE to sample, centrifuge at 13,300 RPM 6. Discard flow through, centrifuge at 13,300 RPM 7. Add 15µL warmed buffer EB to column. Incubate 4 minutes. 8. Centrifuge at 13,300 RPM. 9. Repeat steps 7 and 8 using eluate rather than EB 10. Quantify via nanodrop

Gel Extractioni Overview This protocol covers gel extraction and purification using silica column purification, and are adapted from the Qiagen min-elute kit protocol. Gel extraction is used to purify DNA of a specific length. This protocol assumes that you are using a gel extraction tool and a blue light transluminator, although some steps reference alternatives without going into detail. If using UV light instead of blue light, be sure to take appropriate precautions including appropriate protective equipment. Centrifuge RPMs represent the values of the BEL 24 sample microcentrifuge (Thermofisher).

Background Occasionally, DNA of a specific length will need to be purified from a sample containing multiple different lengths/pieces of DNA. For example, gel extraction is frequently used during restriction enzyme based cloning to isolate sticky ended insert DNA without any of the digested or undigested vectorii. Less commonly, gel extraction is used to isolate a correct length band from a polymerase chain reaction with off target or non-specific amplification.

The fundamental principle of gel extraction is that using electrophoresis, different size bands can be separated on agarose gels. Then the gel can be visualized preferably with blue lightiii and the correct size band can be excised, the agarose dissolved, and the DNA purified.iv Importantly, gel extraction has a reputation in labs as being “voodoo”, i.e. everybody has a different protocol and nobody is sure what works the best or what works decently. On the upside, most protocols that need gel purified inserts (Gibson Assembly, ligations), are quite robust to purity concerns as well as to low concentrations.

Materials - Agarose o Ideally of high quality. Do not use low melt agarose. - 6x Purple Gel Loading Dye No SDS (NEB) o Regular dye (with SDS) is appropriate for use with UV light, but no SDS is preferred (but not required) for use with Syber Safev - TAE Buffer - Ethidium Bromide (ETBR) or Syber Safe (or equivalent) - Diluted 1kb+ ladder (NEB) i Author: Ethan M Jones ii Digested vector has sticky ends which can re-ligate to the insert to form the original plasmid, and undigested vector already is just the original plasmid. iii Blue light is preferred as it does not degrade DNA quality (via double -stranded breaks/cross linkage). When using UV light, exposure time must be minimized, leading to rushing and potentially thicker and more inaccurate gel slices, resulting in lower quality and yield. Additionally, blue light is safer. iv This also has the side effect of removing protein contaminants as well. v This is also true with GelRed. Presumably the SDS interferes with the fluorescence. o Composed of: 1x loading dye and 0.08333 µg/mL ladder - DNA Sample - Isopropanol (100%) - Promega Gel x-tracta or a razor blade - Buffer QG o Used to dissolve the agarose gel slice o Composed of: 5.5 M guanidine thiocyanate, 20mM Tris-HCl, pH 6.6, pH indicatorvi o Note: Do not add bleach to any waste containing buffer QG. Toxic gases will be released. - Buffer PE o Used to wash salts and other non DNA from the column o Composed of: 10mM Tris-HCl pH 7.5, 80% v/v ethanol - Buffer EB (Elution Buffer) o Used to elute DNA from column o Comprised of: 10 mM Tric-HCl pH 8.5 - Blue EconoSpin spin column (with lid)vii - 2.0mL Microcentrifuge Tubes o Temporary tube - 1.7mL Microcentrifuge Tubes o Final tube Equipment: - Gel Box and Lid - Microcentrifuge - Blue light or UV gel imagerviii. - Blue blocking (orange) glasses or UV protective glasses - Heat block

Procedure 1. Prepare a 1%ix agarose gel with the thick side of a 6 well comb.

vi This is almost certainly the same thing as pH indicator I which is used at a 1:250 dilution in buffer PB. Since it transitions from yellow to orange/violet at pH 7.5, it is fairly likely that it is the indicator is just phenol red. However, since we never find the indicator relevant, this buffer can be homemade without its use. vii These give comparable yields to the Qiagen min-elute (purple) columns, at a much lower price. We have found (n = ~5) that the min-elute columns from epoch have lower yields than the regular columns. Presumably this is due to their lower capacity for DNA. [This could potentially be a general principle.] viii Ethidium bromide and Syber Safe can both be used with blue light, however Syber Safe is brighter. UV light is only compatible with Ethidium Bromide (excepting extremely large amounts of DNA). ix Lower gel percentages can lead to better yields/higher purity, however band separation might not be optimal. Going above 1% can lead to better band separation for small fragments, but is not recommended unless required (you’ll know). o Note that you will be able to purify 2 samples per gel. o Take extra care here to make a good gel (dissolve agarose completely) and to use a gel box that works well. o Either ETBR or Syber Safex can be used (Syber Safe is preferred due to increased brightness). o For larger fragments (>3kb) 0.8% Agarose works well. 2. Add 6x loading dye to sample tube (e.g. 10µL 6x loading dye to 50µL of product). o Note that when using Syber Safe the no SDS loading dye should be used. 3. Load 12µLxi of ladder in lane 1 of gel. 4. Load entire sample + dye into the gel. o It is recommended that you put one lane of space between each well with DNA. This prevents cross contamination o It is also recommended that the gel is loaded in the backroom to prevent the contents of the well from spilling out, i.e movement of the gel once loaded should be minimal. 5. Run gel for long enough to produce quality separation. o 30 minutes at 160 volts typically works well 6. Bring Gel(s) to ChemiDoc MP imager. Insert blue tray. Set gel imager to Syber Green for the blue tray (regardless of whether or not you are using Syber Green). Image gel. 7. Determine which bands are the correct ones (to be isolated). 8. Label a 2.0mL temporary tube for each band to be isolated. 9. Preheat heat block to 50C 10. Put on orange glasses and pull tray out. Turn on the transilluminator. Locate the correct size bands. o In some cases (especially with ETBR or faint bands) lab lights will need to be turned off. 11. Either zero the balance with the empty temp tube that will hold the gel, or zero the balance with an empty gel x-tracta tool. (This is done so that the mass of the gel punch may be measured).

x Currently (190303) we use a Syber Safe equivalent fron Genesee. xi This value is not optimized and it more or less may be appropriate. 12. Use the gel x-tractaxii to obtain each band. Note that you will have to perform two punches with for each band (use one x-tracta and do both punches before squeezing). o To use gel x-tracta, place x-tracta on one edge desired band, gently push straight down (wiggling if necessary), pull out at slight angle (20-40 degrees), perform another punch on the remaining part of the band (getting as little extra agarose as possible), place gel x-tracta in tube and squeeze hard and fast. o Remember to get the least amount of agarose without DNA as possible.xiii o When expelling agarose plug, make sure to squeeze hard and fast. 13. With the extracted gel either in the x-tracta tool or in the temp tube (depending on what the balance was zeroed with), measure the mass of the gel punch. 14. Based on the mass of the gel punch, add an amount of buffer QG, in µL, equal to 3 times the mass of the gel slice in milligrams. (i.e. a gel that weighs 0.2g/200mg will receive 600uL of QG). 15. Vortex briefly. 16. Incubate at 50Cxiv for 10 minutes. Vortex every 2-3 minutes. 17. While sample(s) incubate, label a spin column 18. Ensure that the agarose is completely dissolved. The tube should be entirely liquid like and not look thick or ripple when inverted 19. After dissolution, check that the color of the solution is correct. It should be yellow. If it orange or violet, add 10µL 3M sodium acetate (pH 5.0)xv. 20. Add a volume of Isopropanol (in uL) equivalent to the mass of the gel punch (in mg) pipetting to mix. (For example, a gel punch that weighs 0.2g (200mg) will get 200uL of isopropanol). o Once Isopropanol has been added, do not centrifuge sample.

xii Alternatively, use a razorblade. If using a razor blade, be sure to trim the gel as much as possible. xiii This prevents agarose contamination downstream and also limits the need to use large amounts of buffer QG to dissolve the gel slice (QG is the major downstream contaminate in this gel purification protocol) xiv The temperature given is 42-50C. Anecdotally some people swear that melting by hand or in a pocket (i.e. 37C is better). xv Some people swear that performing this step every time improves yields. [Of note, I have never actually performed this step, nor met someone who has performed it] 21. Apply 700µL of sample(s) to column(s). Spin through at 6,000 RPMxvi for 1 minute. Discard flow through (dump out fluid from lower collection tube, then replace the upper column). o Some have said that results are better if flow through is pipetted out rather than dumped.xvii 22. Apply remaining portion of sample to column. Spin through at 6,000 RPM for 1 minute. Discard flow through. 23. Add 750µL buffer PE to column. Remove column from collection tube. Invert column a few times to wash Buffer QG from the walls of the tube. 24. Incubate 5 minutesxviii. Spin through at 13,300 RPMxix for 1 minute. Discard flow through. 25. Add 750µL buffer PE. Remove column from collection tube. Invert column a few times to wash Buffer QG from the walls of the tube. Spin through at 13,300 RPM for 1 minute. Discard flow through. 26. Add 750µL buffer PE. Remove column from collection tube. Invert column a few times to wash Buffer QG from the walls of the tube. Spin through at 13,300 RPM for 1 minute. Discard flow through.xx 27. Preheat aliquot of buffer EB to 50C in heat block. 28. Dry spin at 13,300 RPM for 1 minute to dry sample(s). If significant flow through, discard flow through and dry again. o It is important to discard flow through before performing dry spin. 29. Transfer column(s) to final tubes. Apply 15µL of pre-warmed elution buffer to the column(s). Incubate at room temperature for 4 minutes. Spin through. o Make sure to apply buffer to the membrane and not to the sides of the column 30. (Optional) for increased (~15%) yield, run eluate through column again.

xvi Anecdotal evidence is given on openwetware that this reduction in speed increases yield (due to longer DNA binding time). Alternatively/additionally, one could allow the sample to sit on the column for a minute before spinning. xvii Presumably the important thing is the complete removal of buffer QG xviii This step (and the inversion steps) [allegedly] greatly reduce QG buffer carry over in the purified product. Some have said that performing 5 minute incubation step for the 2nd and even 3rd wash may improve purity. Personally my thoughts are that a 1 minute incubation is likely more than enough, and that performing 1 minute washes 3 times might lead to better results xix The 13,300 RPM spin steps in this protocol are maintained from the original Qiagen protocol. It should not matter if you accidentally use 6,000 RPM instead. xx Adaption from original Qiagen protocol (via ADH), this additional wash has been found to increase purity. 31. Nanodrop to quantify.xxi 32. Store at 4C for use on the same day, and -20C (gel extractions box) for longer term storage.

xxi If DNA is present but impure, a PCR purification may clean up the sample (this is not recommend in general). Abbreviated Protocol 1. Prepare a 1% agarose gel with the thick side of a 6 well comb using Syber Safe. 2. Add 6x loading dye to sample(s) and load gel, run until target band(s) appropriately separated 3. Preheat heat block to 50C. 4. Isolate band(s) using gel x-tracta or razor blade, weigh and record mass of gel slice(s) 5. Add 3µL of Buffer QG per 1mg of gel slice to each gel slice. Vortex 6. Incubate at 50C for 10 minutes, vortex occasionally. 7. Check for complete dissolution of gel slice, and that color is yellow (see full protocol). 8. Add 1µL of isopropanol per 1mg of gel slice. 9. Apply 700µL of sample to spin column. Spin for 1 minute at 6,000 RPM, discard flow through 10. Repeat step 9 until entire sample has been spun through. 11. Preheat buffer EB in heatblock at 50C 12. Add 750µL buffer PE to column. Invert (with lid closed) to wash buffer QG from walls. 13. Incubate 5 minutes. 14. Spin for 1 minute at 13,300 RPM, discard flow through 15. Add 750µL buffer PE to column. Invert (with lid closed) to wash buffer QG from walls. 16. Spin for 1 minute at 13,300 RPM, discard flow through 17. Add 750µL buffer PE to column. Invert (with lid closed) to wash buffer QG from walls. 18. Spin for 1 minute at 13,300 RPM, discard flow through 19. Dry spin for 1 minute at 13,300 RPM 20. Transfer spin column to final tube. Add 15µL buffer EB to column. Incubate 1 minute. Spin for 1 minute at 13,300 RPM 21. Repeat step 19 using eluate 22. Nanodrop for quantification

DpnI Digestioni

Overview This protocol covers the use of the DpnI to remove plasmid template DNA from a PCR reaction. This protocol assumes the use of PCR tubes and a thermocycler, but can easily be altered to use a Heat block or waterbath.

Background The DpnI restriction enzyme digests methylated DNAii-- this serves to remove the template DNA from a tube containing PCR product. This removal of the template DNA reduces background transformation of template plasmid, allowing for more on target assemblies. Note that this process is not recommended or required for purification of restriction enzyme digests/Gel purification products.

Materials - Product to be digested (typically PCR product 24µl in volume) - DpnI (NEB) o As with all restriction enzymes, keep in the freezer (-20C )until immediately before use. Keep in Stratacooler when not in freezer. Enzyme is stored in glycerol (50%) so it should not freeze. - Cutsmart 10x Buffer (NEB) o Comprised of (1x): 50mM Potassium Acetate, 20mM Tris-acetate, 10mM Magnesium Acetate, 100µg.mL BSA, pH 7.9 o This buffer is stored at -20C but an aliquot is usually stored at 4C.

Equipment - Thermocycler or Heat block - Picofuge

Procedure 1. Calculate the correct volume of 10x Cutsmart to add to PCR product tube so that the final concentration of cutsmart is 1x. (Typically 2.7µl 10x Cutsmart to a reaction with 0.5µL of DpnI and 24µL of PCR product [25µL PCR with 1µL used for gel]) o 0.5µL of DpnI is more than enoughiii for a 25µL PCRiv.

i Author: Ethan M Jones ii Specifically GA(methyl)T(opposite base methylated)C iii Smaller volumes would just as well, but are too difficult to pipette. iv Consider the input of plasmid DNA into your PCR (likely 1nM/a few ngs) 2. Add correct amount of 10x Cutsmart to each PCR tubev. o Typically 2.7µL for .5µL DpnI and 24µL PCR product o No need to change to a new PCR tube 3. Add correct amount of DpnI to each tube (typically 0.5µL) o Due to glycerol in storage buffer, enzyme should not make up more than 10% the total volume of reactionvi (e.g. maximum of 2.5µl DpnI in 25µl reaction). 4. Flick to mix then, then spin down in picofuge. 5. Place in thermocyclervii. Heat at 37C for 1 hour, heat inactivate by heating at 80C for 20 minutes, hold at 4C. o Reactions can be incubated at 37C for up to 8 hoursviii, for digestion of larger amounts of DNA. Alternatively, larger volumes of DpnI can be used.ix o Heat inactivation step can be skipped if reaction is immediately PCR purified, as this removes the enzyme anyway. This can be useful to save time.

v Note that DpnI + cutsmart is tolerant of Q5 reaction buffer, but not all restriction enzymes will be. A complete list of tolerance can be found on NEB’s website. vi Technically, the enzyme is stored in 50% glycerol, and no more than 5% of final reaction volume should be glycerol. Simplified in main text for ease of reading. vii Hypothetically this could be done in a heat block with beads (or in a 1.7mL centrifuge tube) viii Via, NEB’s guidelines for the incubation time of various restriction enzymes ix Total restriction enzyme activity is a product of time and enzyme concentration. Although DpnI is fairly inexpensive, all enzymes are fairly costly. However, wasted time is an even greater cost (assuming you do valuable work). In most cases, you will be forced to use vastly too much enzyme relative to your target.

Abbreviated Procedure: 1. Add 10x Cutsmart into PCR product tube to a final concentration of 1x (2.7µL for 24µL PCR product + 0.5µL DpnI) 2. Add 0.5µL DpnI, flick and spin. 3. Incubate for 1 hour at 37C, then incubate for 20 minutes at 80C, then store/hold at 4C. 4. For Gibson pipeline, proceed to PCR purification